AI Readiness Assessment
https://www.bat.com
Assessed2026-07-31
Reference0731/BAT/05
IndustryTobacco and nicotine
Scale187 / 1,968
TrafficNAAwaiting live domain connection, provided as part of service deployment.
Regulated sectorYesTobacco and nicotine sector; no explicit regulator statement in the sampled pages.
Listed companyYes (public)
Manual equivalent15 days · £9,180estimate only
Registered identity from official sources
Registered nameBritish American Tobacco Plc
Country of incorporationUnited Kingdom
Company number03407696
Legal entity identifier213800FKA5MF17RJKT63
Listing
Primary listingLondon Stock Exchange · BATS (XLON)
Also listedNew York Stock Exchange · BTI (ADR)
ISINGB0002875804
SEDOL · MiFIR0287580 · SHRS
SecurityOrdinary Shares of 25p each; fully paid
Trading currencyGBX
Industry codeSIC 2111 · Cigarettes
FCA Official List (Equity shares (commercial companies), LSE) · London Stock Exchange, SETS list (trading identifiers as at 10 February 2020) · GLEIF (legal identity, matched on ISIN) · SEC EDGAR (SIC 2111, Cigarettes) · OpenFIGI (ADR on US) · retrieved 2026-07-31 19:35 UTC
Foundations assessed now · the automated tests
Your Digital Doorway
PeopleAI systems
AI readiness · Pages
29.2%
AI readiness · PDFs
37.6%
Misinformation · Pages
Extremely likely
Misinformation · PDFs
Extremely likely
0%25%50%75%100%
Governancestage 2 · not assessed yet...more
64 judgements need a named person in the organisation to confirm, because they turn on what you know to be true rather than anything measurement can settle: whether a claim is supported, whether a page is still current, which of two versions is the one to quote. In order, they cover: who is answerable for this content; is it still true today; can each claim be backed; does our published content agree; will the right meaning come across; can someone reach it and act on it. An answer recorded against each, with a name and a date, is what turns published content into governed content.
Visibilitystage 3 · not assessed yet...more
Once the foundations are sound and the content is governed, the visibility stage measures how AI systems and agents actually find the estate: which pages they reach, how often, and how the organisation is represented when they answer questions about it, tracked over time. Building visibility on unconfirmed foundations only amplifies whatever is wrong, which is why it comes last.
Key findings · AI readiness
How much of your content is reachable and readable, rather than blocked, buried or missing. Higher is better.
AI readiness is low.the content can be reached, but the way the pages are built keeps much of it from being read cleanly.
What is holding it back
- The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted.34 pages
- A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one.36 pages
- Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text.36 pages
- Without that map the text comes out as the wrong characters, so words and figures are wrong rather than missing.3 documents
- With no headings to divide it the document reads as one undivided block, so a passage loses the section it sat in.4 documents
Key findings · Misinformation
Where your content could lead an AI system to state something incorrect. Only direct routes to a wrong answer count here.
Misinformation risk is high.The pages read as extremely likely and the documents as extremely likely to cause it, which is why the risk reads as high.
What could cause it
- Duplicates collapse to a single surviving version, and which one survives is not the publisher's choice.pages
- When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over.pages
- Without headers a figure is extracted as a bare number, so it gets quoted against the wrong heading.pages
- A figure that comes loose from its heading is read against the wrong row or the wrong column, so the number an AI system quotes is not the number the document gives.documents
- Data living only inside a chart picture is absent from the document's text, so the figures behind it go unstated.documents
- Without that map the text comes out as the wrong characters, so words and figures are wrong rather than missing.documents
▸Overall findings what makes up the figures · click to view
The site as a whole · 10.8%the estate's shared machinery measures 89.7% built. The figure shown is lower because the assessment found repeated pages, boilerplate-dominated near-duplicate pages, which raise the chance of an answer being wrong, and a direct route to a wrong answer caps what can be relied on, however well built. The evidence for each finding is under Site infrastructure below.
Pages · 29.2%the average of 37 pages, each scored on its own build and its own misinformation exposure; today they range 9.6% to 54.7%. Each page's evidence is under Pages considered below.
Documents · 37.6%the average of 5 documents, scored the same way; today they range 9.6% to 57.4%. Each document's evidence is under PDF documents below.
▸Digital doorway measured at the front door · click to view
Anyone asking for this site, a person on a phone or an AI system building an answer, is either let in or turned away, in milliseconds, by machinery most organisations never see. An AI system refused here gets nothing at all: to that system the organisation simply does not answer.
The door changes without notice: a security setting, a network change or a new crawler policy can close it silently. It is measured live on the day of assessment, and it is worth watching continuously rather than once.
People
4 out of 4 served
AI systems
15 out of 15 served
People are served in full, and so are the 15 AI systems.
Whether the estate can be reached and read at all: access rules, links, redirects, sitemap, transport.
▸The site as a whole 10.8% Extremely likely 5 out of 25
| Check | Result | Evidence | |||||||
|---|---|---|---|---|---|---|---|---|---|
| DS16 | fail | 1 URL group(s) are reachable under more than one form (case, trailing slash, or query variants) | |||||||
What we check The same page served at several addresses, varying by capitals, slashes or query strings. What good looks like Each page is served at a single address, not several varying by capitals, slashes or query strings. Why it matters The site cedes the choice of which address is stored and cited, so links and credit scatter across the variants. What we found on this page 1 URL group(s) are reachable under more than one form (case, trailing slash, or query variants) The evidence, taken from the page as served The URLs that serve the same page under more than one form
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: each page is served at a single address, not several varying by capitals, slashes or query strings. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||
| MX12 | fail | 1 pair(s) of pages carry near-identical text (at least 90% of the same words) | |||||||
What we check Distinct URLs among the assessed pages that carry near-identical body text. What good looks like No two assessed pages carry near-identical body text. Why it matters Duplicates collapse to a single surviving version, and which one survives is not the publisher's choice. What we found on this page 1 pair(s) of pages carry near-identical text (at least 90% of the same words) The evidence, taken from the page as served Each page pair whose visible text is near-identical (at least 90% shared vocabulary)
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: no two assessed pages carry near-identical body text. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||
| CT34 | fail | 16 page(s) read as near-identical to other pages in the text a machine ingests (up to 96% shared), but the overlap is shared navigation and footer rather than content: the worst affected page has 34 words of its own out of 623. Distinct pages can then be taken as duplicates and the real content passed over; the pages are listed in the view. | |||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page 16 page(s) read as near-identical to other pages in the text a machine ingests (up to 96% shared), but the overlap is shared navigation and footer rather than content: the worst affected page has 34 words of its own out of 623. Distinct pages can then be taken as duplicates and the real content passed over; the pages are listed in the view. The evidence, taken from the page as served Pages whose own words are lost in a shared navigation and footer (near-identical to a machine)
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||
| DS12 | fail | 15% of the links to sources on other sites did not open when they were checked. 5 were refused, meaning the page is still published but the other site turns away requests that do not come from a person using a browser. That is 5 of the 20 checked, which stand for the 130 links to other sites across the pages checked | |||||||
What we check The share of links pointing off the site, from a sample, whose targets no longer resolve. What good looks like Sources the page relies on open for any reader. Where one refuses or has gone, the page carries the fact in its own words, shows the reference without making it a link, or points somewhere that answers. A refusal by another site is still this site's to resolve, because this site chose to rest a claim on it. Why it matters A source the page leans on did not answer when it was asked for. The claim is left standing with nothing behind it, so the same fact gets taken from elsewhere, the claim is set aside as unsupported, or a partial account is presented as a complete one. Where the source is still published but declined the request, the page still reads as though the evidence sits one step away, so the gap is invisible. What we found on this page 15% of the links to sources on other sites did not open when they were checked. 5 were refused, meaning the page is still published but the other site turns away requests that do not come from a person using a browser. That is 5 of the 20 checked, which stand for the 130 links to other sites across the pages checked The evidence, taken from the page as served The external links checked, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: sources the page relies on open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 5 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||
| CT07 | warning | 1 page title(s) are used on more than one page: “British American Tobacco - Home”; the pages are listed in the view | |||||||
What we check Any title string that is shared by more than one of the assessed pages. What good looks like Each assessed page carries its own distinct title. Why it matters Pages sharing one title carry the same label, so nothing distinguishes them at the point one is picked. What we found on this page 1 page title(s) are used on more than one page: “British American Tobacco - Home”; the pages are listed in the view The evidence, taken from the page as served The pages that share a title, grouped by the title they share
| |||||||||
Passing checks are omitted here; the full record is in the exports.
Pages considered 37 pages
The highest-priority pages, chosen from the site's own signals of importance.
▸https://www.bat.com 53.9% Likely 14 out of 60
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 5 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 5 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 British American Tobacco H2 A Better Tomorrow™: Our purpose H2 A Better Tomorrow™ through strategy & action H2 A Better Tomorrow™ through science & research H2 A Better Tomorrow™ through brands & innovation H2 A Better Tomorrow™ through sustainability H3 Our Chief Executive Tadeu Marroco shares his views on BAT’s Half-Year Results fo H3 Our transformation H2 Omni™ – Forward Thinking for a Smokeless World H2 New Categories fuelling faster transformation H2 Advancing Sustainability for A Better Tomorrow™ H2 People and culture H3 Latest stories and features H4 Video: Chief Executive's take on Half-Year Results 2026 H4 The Smokeless Word: The Rewind - A Look Back at The Smokeless Word So Far H3 Latest press releases H4 Half-Year Report for the six months to 30 June 2026 H4 Creating a stronger BAT through Fit2Win transformation programme H3 In the spotlight H4 2026 Half-Year Results H2 2026 Half-Year Results H2 2026 Half-Year Results H4 Bite-sized science - Vuse level jumped H2 Bite-sized science - Vuse H2 Bite-sized science - Vuse H4 Sustainability leadership level jumped H2 Sustainability leadership H2 Sustainability leadership H4 Tomorrow can't wait level jumped H2 Tomorrow can't wait H2 Tomorrow can't wait H4 Our purpose: To build A Better Tomorrow™ level jumped H2 Our purpose: To build A Better Tomorrow™ H2 Our purpose: To build A Better Tomorrow™ H3 Latest results, reports and research H2 Results centre H2 Annual reporting H2 Omni™ – Forward Thinking for a Smokeless World. H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT04 | warning | the page shows the year 2028 in its text, but nothing on the page states it as the date of the content | |||||||||||
What we check The page carries a date, ideally written into the markup rather than sitting only in the text. What good looks like The page carries a date, ideally written into the markup rather than sitting only in the text. Why it matters An undated page gives nothing to weigh its age against, so old and new material are treated alike. What we found on this page the page shows the year 2028 in its text, but nothing on the page states it as the date of the content The evidence, taken from the page as served The head of this page, as served
The head of the page was read for a tag stating the date of the content. The year shown above was found in the page's own text. <head> <meta name="content-page-ref" content="jKrFtQvSUub5NcsH4cYdxCZhrYfIyWaitb2FGCAy3R23tPBhOzycNAOkqXs-k3au"/> <script defer="defer" type="text/javascript" src="https://rum.hlx.page/.rum/@adobe/helix-rum-js@%5E2/dist/rum-standalone.js" data-routing="env=prod,tier=publish,ams=British American Tobacco"/> <script src="https://cdn.cookielaw.org/scripttemplates/otSDKStub.js" async="" data-domain-script="5fc350fc-73ff-46f2-9463-37d2cf7f5f1a" charset="UTF-8"/> <script> (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push ({'gtm.start': new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0], j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src= 'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f); })(window,document,'script','dataLayer','GTM-TMH7W3WX');</script> <meta charset="UTF-8"/> <title> British American Tobacco - Home </title> <meta name="keywords" content="transforming tobacco,human rights,BAT,Sustainability,British American Tobacco,global company,Investors,reduced risk,Tobacco,annual report,media,features,reporting,next generation products,news"/> <meta name="description" content="The home of BAT, a leading multi-catego… | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-e6c9117394" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-f67a90a4ab" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-6dcf54ecda" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 3 link wording(s) are used for more than one destination, so the same words point at different pages: “find out more”, “read more”, “visit microsite”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 3 link wording(s) are used for more than one destination, so the same words point at different pages: “find out more”, “read more”, “visit microsite”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="corp-hero-carousel__teaser__action-link" href="/who-we-are" target="_blank" rel="noopener"> Find out more </a> <a class="corp-hero-carousel__teaser__action-link" href="/strategy-and-purpose" target="_blank" rel="noopener"> Find out more </a> <a class="corp-hero-carousel__teaser__action-link" href="/science-and-research" target="_blank" rel="noopener"> Find out more </a> <a class="corp-hero-carousel__teaser__action-link" href="/brands-and-innovation" target="_blank" rel="noopener"> Find out more </a> <a class="corp-hero-carousel__teaser__action-link" href="/sustainability-and-esg" target="_blank" rel="noopener"> Find out more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-dd73cbcc2f-cta-ea894a8d3e" rel="noopener noreferrer" aria-label="Visit microsite" href="https://www.asmokelessworld.com/gb/en?utm_source=bat_com&utm_medium=referral&utm_campaign=omni_launch_2024&utm_content=homepage_science" target="_blank">Visit microsi… <a class="cmp-teaser__action-link " id="batcom-teaser-c3e9d99f23-cta-1f44718bb8" aria-label="Read more" href="/brands-and-innovation/driving-innovation">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-e4abb72e3a-cta-2e238cb061" aria-label="Read more" href="/sustainability-and-esg/climate">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-c4fced547c-cta-e72cb5109c" aria-label="Find out more" href="/who-we-are/our-people">Find out more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-cbafa6ba02-cta-1d45dd069b" aria-label="Read more" href="/investors-and-reporting/results-centre">Read more</a> Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 5 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 5 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 British American Tobacco H2 A Better Tomorrow™: Our purpose H2 A Better Tomorrow™ through strategy & action H2 A Better Tomorrow™ through science & research H2 A Better Tomorrow™ through brands & innovation H2 A Better Tomorrow™ through sustainability H3 Our Chief Executive Tadeu Marroco shares his views on BAT’s Half-Year Results fo H3 Our transformation H2 Omni™ – Forward Thinking for a Smokeless World H2 New Categories fuelling faster transformation H2 Advancing Sustainability for A Better Tomorrow™ H2 People and culture H3 Latest stories and features H4 Video: Chief Executive's take on Half-Year Results 2026 H4 The Smokeless Word: The Rewind - A Look Back at The Smokeless Word So Far H3 Latest press releases H4 Half-Year Report for the six months to 30 June 2026 H4 Creating a stronger BAT through Fit2Win transformation programme H3 In the spotlight H4 2026 Half-Year Results H2 2026 Half-Year Results H2 2026 Half-Year Results H4 Bite-sized science - Vuse level jumped H2 Bite-sized science - Vuse H2 Bite-sized science - Vuse H4 Sustainability leadership level jumped H2 Sustainability leadership H2 Sustainability leadership H4 Tomorrow can't wait level jumped H2 Tomorrow can't wait H2 Tomorrow can't wait H4 Our purpose: To build A Better Tomorrow™ level jumped H2 Our purpose: To build A Better Tomorrow™ H2 Our purpose: To build A Better Tomorrow™ H3 Latest results, reports and research H2 Results centre H2 Annual reporting H2 Omni™ – Forward Thinking for a Smokeless World. H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-c3e9d99f23-cta-1f44718bb8" aria-label="Read more" href="/brands-and-innovation/driving-innovation">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-e4abb72e3a-cta-2e238cb061" aria-label="Read more" href="/sustainability-and-esg/climate">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-cbafa6ba02-cta-1d45dd069b" aria-label="Read more" href="/investors-and-reporting/results-centre">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-28c09efc08-cta-a5f1314249" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| AR23 | warning | the video carries captions, but no written transcript sits on the page, so the spoken content cannot be read or quoted as ordinary text | |||||||||||
What we check Whether a written transcript of the video's spoken content sits on the page, as ordinary text a reader can take, separate from any captions on the video itself. What good looks like A transcript linked on the page. A WARNING means the video carries captions but the page has no transcript. When neither captions nor transcript exist, the captions check carries that total absence alone, so one fault is never marked twice. Why it matters Captions live inside the player; a transcript lives on the page, where the spoken words can be read, searched and quoted like any other content. What we found on this page the video carries captions, but no written transcript sits on the page, so the spoken content cannot be read or quoted as ordinary text The evidence, taken from the page as served The video(s) on the page, as served, with no transcript link alongside
<video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/HYR26-Highlights.m3u8" poster="/content/dam/batcom/global/homepage/HYR26-Video-Poster-Frame.jpg" preload="none" playsinline=""> <track src="/content/dam/batcom/global/video-files/main-site/home/HYR26-Highlights.vtt" kind="subtitles" srclang="en" label="English"/> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/VuseAnimation.m3u8" poster="/content/dam/batcom/global/video-files/main-site/bite-sized-science---vuse/VuseAnimation.jpg" preload="none" playsinline=""> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/21YearsInTheDJSI.m3u8" poster="/content/dam/batcom/global/video-files/main-site/sustainability-leadership/20YearsInTheDJSI.jpg" preload="none" playsinline=""> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/tomorrow-cant-wait.m3u8" poster="/content/dam/batcom/global/video-files/main-site/home/tomorrow-cant-wait.jpg" preload="none" playsinline=""> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/OurPurpose2024.m3u8" poster="/content/dam/batcom/global/video-files/main-site/home/OurPurpose.jpg" preload="none" playsinline=""> </video> | |||||||||||||
| MS16 | warning | 2 media resource(s) are named as a technical code, not human-meaningful words, so a machine cannot tell what the resource is from its name | |||||||||||
What we check The name a video, audio clip or image is served under, and whether it is human-meaningful words or a technical code. What good looks like Video, audio and image files carry names made of words that say what they are, not a technical code. Why it matters A resource named as a code or a bare number says nothing about what it is, so a machine cataloguing it has only the code to go on. What we found on this page 2 media resource(s) are named as a technical code, not human-meaningful words, so a machine cannot tell what the resource is from its name The evidence, taken from the page as served The media resources whose name is a technical code, as served
A meaningful name (words that say what it is) lets a machine identify the resource; a code or a bare number cannot. <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/21YearsInTheDJSI.m3u8" poster="/content/dam/batcom/global/video-files/main-site/sustainability-leadership/20YearsInTheDJSI.jpg" preload="none" playsinline=""> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/OurPurpose2024.m3u8" poster="/content/dam/batcom/global/video-files/main-site/home/OurPurpose.jpg" preload="none" playsinline=""> </video> | |||||||||||||
| AR03 | warning | 18% of the words a person sees are missing from the page as first delivered (8889 characters delivered, 12552 once the page finishes building itself) | |||||||||||
What we check How much of the page's content exists only once JavaScript has run. What good looks like The page's content is present in the served HTML, so it does not depend on JavaScript to appear. Why it matters The page is fetched successfully but the answer is missing from what comes back, so the content is absent rather than wrong. What we found on this page 18% of the words a person sees are missing from the page as first delivered (8889 characters delivered, 12552 once the page finishes building itself) The evidence, taken from the page as served Content present only after JavaScript runs (absent from the served HTML)
The served HTML carries 8889 characters of text; after JavaScript runs the page carries 12552. 18% of the rendered page's words are absent from the HTML the server sends. What the server sends (raw HTML text, first 280 chars) British American Tobacco - Home Skip to main content Media Stories and features Press Releases Subscribe to News Alert Media contacts Social media Contact us Careers Who we are Our business Our business model At a glance Our values Our people Life at BAT Embracing Diversity and I…
Words present only after JavaScript runs
4570ableactiveallowalwaysanalyticsapplyareasbecausebeingbillingblockingbrowsercancelcartcentercheckboxchoicechoicesclearclickcountdirectlydoeseacheasilyembedenableenablesensuringexampleexpectfilterfindingformfunctionalfunctionalitygivehelpshowevericonidentifyincludeinterestintolabellanguagelikelistlookingmakemightmostlymovenumberofferoperationpartyperformancepersonalized
| |||||||||||||
| AR11 | warning | 18% of the words a person sees are missing from the page as first delivered, which is 0.71 of its finished size | |||||||||||
What we check The measured gap between the page as served and the page after the browser has built it. What good looks like The served page and the fully built page carry the same words, with little gap between them. Why it matters Two versions of the same page exist, and which words come back depends on which version was fetched. What we found on this page 18% of the words a person sees are missing from the page as first delivered, which is 0.71 of its finished size The evidence, taken from the page as served This page as first delivered, against the page once its scripts have run
Both were taken from the same request. The first is what the site sent; the second is what a person ends up looking at.
| |||||||||||||
| AR12 | warning | 67% of the page's text (8462 of 12552 characters) is not visible before interaction | |||||||||||
What we check How much of the page's text is hidden behind a click or an expander before anyone interacts. What good looks like The page's text is visible as it loads, not held behind a click or an expander. Why it matters Half or more of the page's text is present but never shown, so what readers see and what it contains diverge. What we found on this page 67% of the page's text (8462 of 12552 characters) is not visible before interaction The evidence, taken from the page as served How much of this page is hidden until something is clicked
Counted on the page once its scripts had run, before anything was clicked.
| |||||||||||||
| MS04 | warning | the page's naming markup is faulty: 9 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 9 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<button class="batcom-video__info-icon" aria-label="Open video info"/> <button class="batcom-video__button-play" type="button" aria-label="Play video" data-nc-play="Play video"/> <button class="batcom-video__info-icon" aria-label="Open video info"/> <button class="batcom-video__button-play" type="button" aria-label="Play video" data-nc-play="Play video"/> <button class="batcom-video__info-icon" aria-label="Open video info"/> <button class="batcom-video__button-play" type="button" aria-label="Play video" data-nc-play="Play video"/> <button class="batcom-video__info-icon" aria-label="Open video info"/> <button class="batcom-video__button-play" type="button" aria-label="Play video" data-nc-play="Play video"/> <button class="batcom-flyout__closebutton">Close <span class="batcom-flyout__closeicon"/> </button> | |||||||||||||
| AJ08 | warning | focus order: the first 12 Tab stops do not follow document order | |||||||||||
What we check Whether pressing Tab moves through the page in the order its content is written. What good looks like Pressing Tab moves through the page in the order its content is written. Why it matters The keyboard jumps around the page out of sequence, so a form is worked through in an order nobody intended. What we found on this page focus order: the first 12 Tab stops do not follow document order The evidence, taken from the page as served How the keyboard moves through this page
Read on the page once its scripts had run, following the keyboard from the top.
| |||||||||||||
▸https://www.bat.com/ 53.7% Likely 13 out of 60
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 5 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 5 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 British American Tobacco H2 A Better Tomorrow™: Our purpose H2 A Better Tomorrow™ through strategy & action H2 A Better Tomorrow™ through science & research H2 A Better Tomorrow™ through brands & innovation H2 A Better Tomorrow™ through sustainability H3 Our Chief Executive Tadeu Marroco shares his views on BAT’s Half-Year Results fo H3 Our transformation H2 Omni™ – Forward Thinking for a Smokeless World H2 New Categories fuelling faster transformation H2 Advancing Sustainability for A Better Tomorrow™ H2 People and culture H3 Latest stories and features H4 Video: Chief Executive's take on Half-Year Results 2026 H4 The Smokeless Word: The Rewind - A Look Back at The Smokeless Word So Far H3 Latest press releases H4 Half-Year Report for the six months to 30 June 2026 H4 Creating a stronger BAT through Fit2Win transformation programme H3 In the spotlight H4 2026 Half-Year Results H2 2026 Half-Year Results H2 2026 Half-Year Results H4 Bite-sized science - Vuse level jumped H2 Bite-sized science - Vuse H2 Bite-sized science - Vuse H4 Sustainability leadership level jumped H2 Sustainability leadership H2 Sustainability leadership H4 Tomorrow can't wait level jumped H2 Tomorrow can't wait H2 Tomorrow can't wait H4 Our purpose: To build A Better Tomorrow™ level jumped H2 Our purpose: To build A Better Tomorrow™ H2 Our purpose: To build A Better Tomorrow™ H3 Latest results, reports and research H2 Results centre H2 Annual reporting H2 Omni™ – Forward Thinking for a Smokeless World. H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT04 | warning | the page shows the year 2028 in its text, but nothing on the page states it as the date of the content | |||||||||||
What we check The page carries a date, ideally written into the markup rather than sitting only in the text. What good looks like The page carries a date, ideally written into the markup rather than sitting only in the text. Why it matters An undated page gives nothing to weigh its age against, so old and new material are treated alike. What we found on this page the page shows the year 2028 in its text, but nothing on the page states it as the date of the content The evidence, taken from the page as served The head of this page, as served
The head of the page was read for a tag stating the date of the content. The year shown above was found in the page's own text. <head> <meta name="content-page-ref" content="jKrFtQvSUub5NcsH4cYdxCZhrYfIyWaitb2FGCAy3R23tPBhOzycNAOkqXs-k3au"/> <script defer="defer" type="text/javascript" src="https://rum.hlx.page/.rum/@adobe/helix-rum-js@%5E2/dist/rum-standalone.js" data-routing="env=prod,tier=publish,ams=British American Tobacco"/> <script src="https://cdn.cookielaw.org/scripttemplates/otSDKStub.js" async="" data-domain-script="5fc350fc-73ff-46f2-9463-37d2cf7f5f1a" charset="UTF-8"/> <script> (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push ({'gtm.start': new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0], j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src= 'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f); })(window,document,'script','dataLayer','GTM-TMH7W3WX');</script> <meta charset="UTF-8"/> <title> British American Tobacco - Home </title> <meta name="keywords" content="transforming tobacco,human rights,BAT,Sustainability,British American Tobacco,global company,Investors,reduced risk,Tobacco,annual report,media,features,reporting,next generation products,news"/> <meta name="description" content="The home of BAT, a leading multi-catego… | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-e6c9117394" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-f67a90a4ab" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-6dcf54ecda" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 3 link wording(s) are used for more than one destination, so the same words point at different pages: “find out more”, “read more”, “visit microsite”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 3 link wording(s) are used for more than one destination, so the same words point at different pages: “find out more”, “read more”, “visit microsite”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="corp-hero-carousel__teaser__action-link" href="/who-we-are" target="_blank" rel="noopener"> Find out more </a> <a class="corp-hero-carousel__teaser__action-link" href="/strategy-and-purpose" target="_blank" rel="noopener"> Find out more </a> <a class="corp-hero-carousel__teaser__action-link" href="/science-and-research" target="_blank" rel="noopener"> Find out more </a> <a class="corp-hero-carousel__teaser__action-link" href="/brands-and-innovation" target="_blank" rel="noopener"> Find out more </a> <a class="corp-hero-carousel__teaser__action-link" href="/sustainability-and-esg" target="_blank" rel="noopener"> Find out more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-dd73cbcc2f-cta-ea894a8d3e" rel="noopener noreferrer" aria-label="Visit microsite" href="https://www.asmokelessworld.com/gb/en?utm_source=bat_com&utm_medium=referral&utm_campaign=omni_launch_2024&utm_content=homepage_science" target="_blank">Visit microsi… <a class="cmp-teaser__action-link " id="batcom-teaser-c3e9d99f23-cta-1f44718bb8" aria-label="Read more" href="/brands-and-innovation/driving-innovation">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-e4abb72e3a-cta-2e238cb061" aria-label="Read more" href="/sustainability-and-esg/climate">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-c4fced547c-cta-e72cb5109c" aria-label="Find out more" href="/who-we-are/our-people">Find out more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-cbafa6ba02-cta-1d45dd069b" aria-label="Read more" href="/investors-and-reporting/results-centre">Read more</a> Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 5 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 5 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 British American Tobacco H2 A Better Tomorrow™: Our purpose H2 A Better Tomorrow™ through strategy & action H2 A Better Tomorrow™ through science & research H2 A Better Tomorrow™ through brands & innovation H2 A Better Tomorrow™ through sustainability H3 Our Chief Executive Tadeu Marroco shares his views on BAT’s Half-Year Results fo H3 Our transformation H2 Omni™ – Forward Thinking for a Smokeless World H2 New Categories fuelling faster transformation H2 Advancing Sustainability for A Better Tomorrow™ H2 People and culture H3 Latest stories and features H4 Video: Chief Executive's take on Half-Year Results 2026 H4 The Smokeless Word: The Rewind - A Look Back at The Smokeless Word So Far H3 Latest press releases H4 Half-Year Report for the six months to 30 June 2026 H4 Creating a stronger BAT through Fit2Win transformation programme H3 In the spotlight H4 2026 Half-Year Results H2 2026 Half-Year Results H2 2026 Half-Year Results H4 Bite-sized science - Vuse level jumped H2 Bite-sized science - Vuse H2 Bite-sized science - Vuse H4 Sustainability leadership level jumped H2 Sustainability leadership H2 Sustainability leadership H4 Tomorrow can't wait level jumped H2 Tomorrow can't wait H2 Tomorrow can't wait H4 Our purpose: To build A Better Tomorrow™ level jumped H2 Our purpose: To build A Better Tomorrow™ H2 Our purpose: To build A Better Tomorrow™ H3 Latest results, reports and research H2 Results centre H2 Annual reporting H2 Omni™ – Forward Thinking for a Smokeless World. H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-c3e9d99f23-cta-1f44718bb8" aria-label="Read more" href="/brands-and-innovation/driving-innovation">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-e4abb72e3a-cta-2e238cb061" aria-label="Read more" href="/sustainability-and-esg/climate">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-cbafa6ba02-cta-1d45dd069b" aria-label="Read more" href="/investors-and-reporting/results-centre">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-28c09efc08-cta-a5f1314249" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| AR23 | warning | the video carries captions, but no written transcript sits on the page, so the spoken content cannot be read or quoted as ordinary text | |||||||||||
What we check Whether a written transcript of the video's spoken content sits on the page, as ordinary text a reader can take, separate from any captions on the video itself. What good looks like A transcript linked on the page. A WARNING means the video carries captions but the page has no transcript. When neither captions nor transcript exist, the captions check carries that total absence alone, so one fault is never marked twice. Why it matters Captions live inside the player; a transcript lives on the page, where the spoken words can be read, searched and quoted like any other content. What we found on this page the video carries captions, but no written transcript sits on the page, so the spoken content cannot be read or quoted as ordinary text The evidence, taken from the page as served The video(s) on the page, as served, with no transcript link alongside
<video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/HYR26-Highlights.m3u8" poster="/content/dam/batcom/global/homepage/HYR26-Video-Poster-Frame.jpg" preload="none" playsinline=""> <track src="/content/dam/batcom/global/video-files/main-site/home/HYR26-Highlights.vtt" kind="subtitles" srclang="en" label="English"/> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/VuseAnimation.m3u8" poster="/content/dam/batcom/global/video-files/main-site/bite-sized-science---vuse/VuseAnimation.jpg" preload="none" playsinline=""> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/21YearsInTheDJSI.m3u8" poster="/content/dam/batcom/global/video-files/main-site/sustainability-leadership/20YearsInTheDJSI.jpg" preload="none" playsinline=""> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/tomorrow-cant-wait.m3u8" poster="/content/dam/batcom/global/video-files/main-site/home/tomorrow-cant-wait.jpg" preload="none" playsinline=""> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/OurPurpose2024.m3u8" poster="/content/dam/batcom/global/video-files/main-site/home/OurPurpose.jpg" preload="none" playsinline=""> </video> | |||||||||||||
| MS16 | warning | 2 media resource(s) are named as a technical code, not human-meaningful words, so a machine cannot tell what the resource is from its name | |||||||||||
What we check The name a video, audio clip or image is served under, and whether it is human-meaningful words or a technical code. What good looks like Video, audio and image files carry names made of words that say what they are, not a technical code. Why it matters A resource named as a code or a bare number says nothing about what it is, so a machine cataloguing it has only the code to go on. What we found on this page 2 media resource(s) are named as a technical code, not human-meaningful words, so a machine cannot tell what the resource is from its name The evidence, taken from the page as served The media resources whose name is a technical code, as served
A meaningful name (words that say what it is) lets a machine identify the resource; a code or a bare number cannot. <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/21YearsInTheDJSI.m3u8" poster="/content/dam/batcom/global/video-files/main-site/sustainability-leadership/20YearsInTheDJSI.jpg" preload="none" playsinline=""> </video> <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/home/OurPurpose2024.m3u8" poster="/content/dam/batcom/global/video-files/main-site/home/OurPurpose.jpg" preload="none" playsinline=""> </video> | |||||||||||||
| AR03 | warning | 22% of the words a person sees are missing from the page as first delivered (8889 characters delivered, 12010 once the page finishes building itself) | |||||||||||
What we check How much of the page's content exists only once JavaScript has run. What good looks like The page's content is present in the served HTML, so it does not depend on JavaScript to appear. Why it matters The page is fetched successfully but the answer is missing from what comes back, so the content is absent rather than wrong. What we found on this page 22% of the words a person sees are missing from the page as first delivered (8889 characters delivered, 12010 once the page finishes building itself) The evidence, taken from the page as served Content present only after JavaScript runs (absent from the served HTML)
The served HTML carries 8889 characters of text; after JavaScript runs the page carries 12010. 22% of the rendered page's words are absent from the HTML the server sends. What the server sends (raw HTML text, first 280 chars) British American Tobacco - Home Skip to main content Media Stories and features Press Releases Subscribe to News Alert Media contacts Social media Contact us Careers Who we are Our business Our business model At a glance Our values Our people Life at BAT Embracing Diversity and I…
Words present only after JavaScript runs
4570ableacceptactiveallowalwaysanalyticsapplyareasbecausebeingbillingblockingbrowserbuttoncancelcartcentercheckboxchoicechoiceschooseclearclickconsentcookiescountdevicedirectlydisabledoeseacheasilyembedenableenablesenhanceensuringessentialexampleexpectexperiencefilterfindingformfunctionalfunctionalitygivehelpshowevericonidentifyimproveincludeinterestintolabellanguagelikelist
| |||||||||||||
| AR11 | warning | 22% of the words a person sees are missing from the page as first delivered, which is 0.74 of its finished size | |||||||||||
What we check The measured gap between the page as served and the page after the browser has built it. What good looks like The served page and the fully built page carry the same words, with little gap between them. Why it matters Two versions of the same page exist, and which words come back depends on which version was fetched. What we found on this page 22% of the words a person sees are missing from the page as first delivered, which is 0.74 of its finished size The evidence, taken from the page as served This page as first delivered, against the page once its scripts have run
Both were taken from the same request. The first is what the site sent; the second is what a person ends up looking at.
| |||||||||||||
| AR12 | warning | 65% of the page's text (7889 of 12010 characters) is not visible before interaction | |||||||||||
What we check How much of the page's text is hidden behind a click or an expander before anyone interacts. What good looks like The page's text is visible as it loads, not held behind a click or an expander. Why it matters Half or more of the page's text is present but never shown, so what readers see and what it contains diverge. What we found on this page 65% of the page's text (7889 of 12010 characters) is not visible before interaction The evidence, taken from the page as served How much of this page is hidden until something is clicked
Counted on the page once its scripts had run, before anything was clicked.
| |||||||||||||
| MS04 | warning | the page's naming markup is faulty: 9 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 9 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<button class="batcom-video__info-icon" aria-label="Open video info"/> <button class="batcom-video__button-play" type="button" aria-label="Play video" data-nc-play="Play video"/> <button class="batcom-video__info-icon" aria-label="Open video info"/> <button class="batcom-video__button-play" type="button" aria-label="Play video" data-nc-play="Play video"/> <button class="batcom-video__info-icon" aria-label="Open video info"/> <button class="batcom-video__button-play" type="button" aria-label="Play video" data-nc-play="Play video"/> <button class="batcom-video__info-icon" aria-label="Open video info"/> <button class="batcom-video__button-play" type="button" aria-label="Play video" data-nc-play="Play video"/> <button class="batcom-flyout__closebutton">Close <span class="batcom-flyout__closeicon"/> </button> | |||||||||||||
▸https://www.bat.com/investors-and-reporting/reporting/omni-report 31.6% Very likely 14 out of 61
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT06 | warning | title length 73 chars | |||||||||||
What we check The title's length sits in a usable window, between 10 and 65 characters. What good looks like The title's length sits in a usable window, neither too short to say anything nor too long to be cut. Why it matters A title outside that window is liable to be rewritten elsewhere, and control of the page's name goes with it. What we found on this page title length 73 chars The evidence, taken from the page as served The page title, as served
<title> British American Tobacco - Omni™ – Forward Thinking for a Smokeless World </title> <title>X (Twitter)</title> <title>Facebook</title> <title>LinkedIn</title> <title>Email</title> <title>X</title> <title>YouTube</title> <title>LinkedIn</title> <title>Glassdoor</title> | |||||||||||||
| CT32 | warning | length 211 | |||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 211 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="The Omni™ is a progress summary of our journey to A Better Tomorrow™ by Building a Smokeless World. A smokeless world built on Smokeless Products where, ultimately, cigarettes have become a thing of the past. "/> | |||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Omni™ – Forward Thinking for a Smokeless World H2 Omni™ – Forward Thinking for a Smokeless World H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Finvestors-and-reporting%2Freporting%2Fomni-report" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Omni™ – Forward Thinking for a Smokeless World H2 Omni™ – Forward Thinking for a Smokeless World H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-505b5987a6-cta-d3dad6a9e2" data-cmp-data-layer="{"batcom-teaser-505b5987a6-cta-d3dad6a9e2":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| AR03 | warning | 26% of the words a person sees are missing from the page as first delivered (5327 characters delivered, 8418 once the page finishes building itself) | |||||||||||
What we check How much of the page's content exists only once JavaScript has run. What good looks like The page's content is present in the served HTML, so it does not depend on JavaScript to appear. Why it matters The page is fetched successfully but the answer is missing from what comes back, so the content is absent rather than wrong. What we found on this page 26% of the words a person sees are missing from the page as first delivered (5327 characters delivered, 8418 once the page finishes building itself) The evidence, taken from the page as served Content present only after JavaScript runs (absent from the served HTML)
The served HTML carries 5327 characters of text; after JavaScript runs the page carries 8418. 26% of the rendered page's words are absent from the HTML the server sends. What the server sends (raw HTML text, first 280 chars) British American Tobacco - Omni™ – Forward Thinking for a Smokeless World Skip to main content Media Stories and features Press Releases Subscribe to News Alert Media contacts Social media Contact us Careers Who we are Our business Our business model At a glance Our values Our pe…
Words present only after JavaScript runs
20264570ableacceptactiveallowalwaysanalyticsapplyareasaroundbackbecausebeingbillingblockingbrowserbuttoncancelcartcentercheckboxchoicechoiceschooseclearclickconsentcookiescountdevicedirectlydisabledoeseacheasilyembedenablesenhanceensuringessentialexampleexpectexperiencefilterfindingformfunctionalfunctionalitygivehelpshowevericonidentifyimpactimproveincludeinterestintolabel
| |||||||||||||
| AR11 | warning | 26% of the words a person sees are missing from the page as first delivered, which is 0.63 of its finished size | |||||||||||
What we check The measured gap between the page as served and the page after the browser has built it. What good looks like The served page and the fully built page carry the same words, with little gap between them. Why it matters Two versions of the same page exist, and which words come back depends on which version was fetched. What we found on this page 26% of the words a person sees are missing from the page as first delivered, which is 0.63 of its finished size The evidence, taken from the page as served This page as first delivered, against the page once its scripts have run
Both were taken from the same request. The first is what the site sent; the second is what a person ends up looking at.
| |||||||||||||
| AR12 | warning | 71% of the page's text (6026 of 8418 characters) is not visible before interaction | |||||||||||
What we check How much of the page's text is hidden behind a click or an expander before anyone interacts. What good looks like The page's text is visible as it loads, not held behind a click or an expander. Why it matters Half or more of the page's text is present but never shown, so what readers see and what it contains diverge. What we found on this page 71% of the page's text (6026 of 8418 characters) is not visible before interaction The evidence, taken from the page as served How much of this page is hidden until something is clicked
Counted on the page once its scripts had run, before anything was clicked.
| |||||||||||||
| MS04 | warning | the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<button class="batcom-flyout__closebutton">Close <span class="batcom-flyout__closeicon"/> </button> | |||||||||||||
| CT34 | fail | the page's own words are thin: 589 of its 655 words are shared navigation and footer, so it reads as 92% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 589 of its 655 words are shared navigation and footer, so it reads as 92% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
▸https://www.bat.com/investors-and-reporting/reporting/combined-annual-and-esg-report 31.9% Very likely 13 out of 61
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT06 | warning | title length 68 chars | |||||||||||
What we check The title's length sits in a usable window, between 10 and 65 characters. What good looks like The title's length sits in a usable window, neither too short to say anything nor too long to be cut. Why it matters A title outside that window is liable to be rewritten elsewhere, and control of the page's name goes with it. What we found on this page title length 68 chars The evidence, taken from the page as served The page title, as served
<title> British American Tobacco - Combined Annual and Sustainability Report </title> <title>X (Twitter)</title> <title>Facebook</title> <title>LinkedIn</title> <title>Email</title> <title>X</title> <title>YouTube</title> <title>LinkedIn</title> <title>Glassdoor</title> | |||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Combined Annual and Sustainability Report H2 Combined Annual and Sustainability Report H3 Combined Annual and Sustainability Report 2025 H4 Combined Performance and Sustainability Summary H4 Annual Report H3 Previous reporting H3 (no text) H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Finvestors-and-reporting%2Freporting%2Fcombined-annual-and-esg-report" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23"… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Combined Annual and Sustainability Report H2 Combined Annual and Sustainability Report H3 Combined Annual and Sustainability Report 2025 H4 Combined Performance and Sustainability Summary H4 Annual Report H3 Previous reporting H3 (no text) H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-d85f9a84a9-cta-d3dad6a9e2" data-cmp-data-layer="{"batcom-teaser-d85f9a84a9-cta-d3dad6a9e2":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| AR03 | warning | 29% of the words a person sees are missing from the page as first delivered (5753 characters delivered, 8844 once the page finishes building itself) | |||||||||||
What we check How much of the page's content exists only once JavaScript has run. What good looks like The page's content is present in the served HTML, so it does not depend on JavaScript to appear. Why it matters The page is fetched successfully but the answer is missing from what comes back, so the content is absent rather than wrong. What we found on this page 29% of the words a person sees are missing from the page as first delivered (5753 characters delivered, 8844 once the page finishes building itself) The evidence, taken from the page as served Content present only after JavaScript runs (absent from the served HTML)
The served HTML carries 5753 characters of text; after JavaScript runs the page carries 8844. 29% of the rendered page's words are absent from the HTML the server sends. What the server sends (raw HTML text, first 280 chars) British American Tobacco - Combined Annual and Sustainability Report Skip to main content Media Stories and features Press Releases Subscribe to News Alert Media contacts Social media Contact us Careers Who we are Our business Our business model At a glance Our values Our people …
Words present only after JavaScript runs
20264570ableaboutacceptactiveallowalwaysanalyticsapplyareasaroundbackbecausebeingbillingblockingbrowserbuttoncancelcartcentercheckboxchoicechoiceschooseclearclickconsentcookiescountdevicedirectlydisabledoeseacheasilyembedenableenablesenhanceensuringessentialexampleexpectexperiencefilterfindingfromfunctionalfunctionalitygivehavehelpshowevericonidentifyimpactimproveinclude
| |||||||||||||
| AR11 | warning | 29% of the words a person sees are missing from the page as first delivered, which is 0.65 of its finished size | |||||||||||
What we check The measured gap between the page as served and the page after the browser has built it. What good looks like The served page and the fully built page carry the same words, with little gap between them. Why it matters Two versions of the same page exist, and which words come back depends on which version was fetched. What we found on this page 29% of the words a person sees are missing from the page as first delivered, which is 0.65 of its finished size The evidence, taken from the page as served This page as first delivered, against the page once its scripts have run
Both were taken from the same request. The first is what the site sent; the second is what a person ends up looking at.
| |||||||||||||
| AR12 | warning | 68% of the page's text (6053 of 8844 characters) is not visible before interaction | |||||||||||
What we check How much of the page's text is hidden behind a click or an expander before anyone interacts. What good looks like The page's text is visible as it loads, not held behind a click or an expander. Why it matters Half or more of the page's text is present but never shown, so what readers see and what it contains diverge. What we found on this page 68% of the page's text (6053 of 8844 characters) is not visible before interaction The evidence, taken from the page as served How much of this page is hidden until something is clicked
Counted on the page once its scripts had run, before anything was clicked.
| |||||||||||||
| MS04 | warning | the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<button class="batcom-flyout__closebutton">Close <span class="batcom-flyout__closeicon"/> </button> | |||||||||||||
| CT34 | fail | the page's own words are thin: 588 of its 628 words are shared navigation and footer, so it reads as 95% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 588 of its 628 words are shared navigation and footer, so it reads as 95% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
▸https://www.bat.com/investors-and-reporting/reporting/sustainability-reporting/people-and-culture-report 32.2% Very likely 12 out of 61
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 People and Culture Report H2 People and Culture Report H3 2025 People and Culture Report H3 Previous Gender Pay reporting H3 (no text) H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Finvestors-and-reporting%2Freporting%2Fsustainability-reporting%2Fpeople-and-culture-report" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg … <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 People and Culture Report H2 People and Culture Report H3 2025 People and Culture Report H3 Previous Gender Pay reporting H3 (no text) H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-04618fd453-cta-d3dad6a9e2" data-cmp-data-layer="{"batcom-teaser-04618fd453-cta-d3dad6a9e2":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| AR03 | warning | 30% of the words a person sees are missing from the page as first delivered (5171 characters delivered, 8262 once the page finishes building itself) | |||||||||||
What we check How much of the page's content exists only once JavaScript has run. What good looks like The page's content is present in the served HTML, so it does not depend on JavaScript to appear. Why it matters The page is fetched successfully but the answer is missing from what comes back, so the content is absent rather than wrong. What we found on this page 30% of the words a person sees are missing from the page as first delivered (5171 characters delivered, 8262 once the page finishes building itself) The evidence, taken from the page as served Content present only after JavaScript runs (absent from the served HTML)
The served HTML carries 5171 characters of text; after JavaScript runs the page carries 8262. 30% of the rendered page's words are absent from the HTML the server sends. What the server sends (raw HTML text, first 280 chars) British American Tobacco - People and Culture Report Skip to main content Media Stories and features Press Releases Subscribe to News Alert Media contacts Social media Contact us Careers Who we are Our business Our business model At a glance Our values Our people Life at BAT Embr…
Words present only after JavaScript runs
20264570ableaboutacceptactiveallowalwaysanalyticsapplyareasaroundbackbecausebeingbillingblockingbrowserbuttoncancelcartcentercheckboxchoicechoiceschooseclearclickconsentcookiescountdevicedirectlydisabledoeseacheasilyembedenableenablesenhanceensuringessentialexampleexpectexperiencefilterfindingformfromfunctionalfunctionalitygivehavehelpshowevericonidentifyimpactimprove
| |||||||||||||
| AR11 | warning | 30% of the words a person sees are missing from the page as first delivered, which is 0.63 of its finished size | |||||||||||
What we check The measured gap between the page as served and the page after the browser has built it. What good looks like The served page and the fully built page carry the same words, with little gap between them. Why it matters Two versions of the same page exist, and which words come back depends on which version was fetched. What we found on this page 30% of the words a person sees are missing from the page as first delivered, which is 0.63 of its finished size The evidence, taken from the page as served This page as first delivered, against the page once its scripts have run
Both were taken from the same request. The first is what the site sent; the second is what a person ends up looking at.
| |||||||||||||
| AR12 | warning | 72% of the page's text (6005 of 8262 characters) is not visible before interaction | |||||||||||
What we check How much of the page's text is hidden behind a click or an expander before anyone interacts. What good looks like The page's text is visible as it loads, not held behind a click or an expander. Why it matters Half or more of the page's text is present but never shown, so what readers see and what it contains diverge. What we found on this page 72% of the page's text (6005 of 8262 characters) is not visible before interaction The evidence, taken from the page as served How much of this page is hidden until something is clicked
Counted on the page once its scripts had run, before anything was clicked.
| |||||||||||||
| MS04 | warning | the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<button class="batcom-flyout__closebutton">Close <span class="batcom-flyout__closeicon"/> </button> | |||||||||||||
| CT34 | fail | the page's own words are thin: 589 of its 623 words are shared navigation and footer, so it reads as 96% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 589 of its 623 words are shared navigation and footer, so it reads as 96% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
▸https://www.bat.com/science-and-research/world-class-science 54.1% Likely 11 out of 60
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 World-class science H2 World-class science H3 Testing our products: our nine-step risk assessment framework H4 A weight of evidence approach to assess risk H5 Emissions H6 Combustion studies H6 Emissions studies H6 Toxicological studies H5 Exposure H6 Use behaviour H6 Clinical: Pharmacokinetic (PK) studies H6 Clinical: Exposure H5 Risk H6 Clinical individual risk H6 Populations risk (Post-marketing surveillance) H6 Epidemiological modelling data H5 Long-term: Epidemiological data H2 (no text) H3 Our commitment H3 The challenges H3 Animal testing H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Science & Research level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fscience-and-research%2Fworld-class-science" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill=… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 World-class science H2 World-class science H3 Testing our products: our nine-step risk assessment framework H4 A weight of evidence approach to assess risk H5 Emissions H6 Combustion studies H6 Emissions studies H6 Toxicological studies H5 Exposure H6 Use behaviour H6 Clinical: Pharmacokinetic (PK) studies H6 Clinical: Exposure H5 Risk H6 Clinical individual risk H6 Populations risk (Post-marketing surveillance) H6 Epidemiological modelling data H5 Long-term: Epidemiological data H2 (no text) H3 Our commitment H3 The challenges H3 Animal testing H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Science & Research level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-29d5fb77a6-cta-51c70db8c9" data-cmp-data-layer="{"batcom-teaser-29d5fb77a6-cta-51c70db8c9":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| AR03 | warning | 12% of the words a person sees are missing from the page as first delivered (13530 characters delivered, 16631 once the page finishes building itself) | |||||||||||||||||||
What we check How much of the page's content exists only once JavaScript has run. What good looks like The page's content is present in the served HTML, so it does not depend on JavaScript to appear. Why it matters The page is fetched successfully but the answer is missing from what comes back, so the content is absent rather than wrong. What we found on this page 12% of the words a person sees are missing from the page as first delivered (13530 characters delivered, 16631 once the page finishes building itself) The evidence, taken from the page as served Content present only after JavaScript runs (absent from the served HTML)
The served HTML carries 13530 characters of text; after JavaScript runs the page carries 16631. 12% of the rendered page's words are absent from the HTML the server sends. What the server sends (raw HTML text, first 280 chars) British American Tobacco - World-class science Skip to main content Media Stories and features Press Releases Subscribe to News Alert Media contacts Social media Contact us Careers Who we are Our business Our business model At a glance Our values Our people Life at BAT Embracing …
Words present only after JavaScript runs
20264570aboutacceptactiveanalyticsapplyareasbackbecausebillingblockingbrowserbuttoncancelcartcentercheckboxchoiceschooseclearconsentcookiescountdevicedirectlydisabledoeseasilyembedenableenablesenhanceensuringexpectexperiencefilterfindingformfunctionalfunctionalitygivehowevericonidentifyimproveincludeinterestintolabellanguagelikelistlookingmostlymovenumberofferoperationpage
| |||||||||||||||||||||
| AR11 | warning | 12% of the words a person sees are missing from the page as first delivered, which is 0.81 of its finished size | |||||||||||||||||||
What we check The measured gap between the page as served and the page after the browser has built it. What good looks like The served page and the fully built page carry the same words, with little gap between them. Why it matters Two versions of the same page exist, and which words come back depends on which version was fetched. What we found on this page 12% of the words a person sees are missing from the page as first delivered, which is 0.81 of its finished size The evidence, taken from the page as served This page as first delivered, against the page once its scripts have run
Both were taken from the same request. The first is what the site sent; the second is what a person ends up looking at.
| |||||||||||||||||||||
| AR12 | warning | 64% of the page's text (10684 of 16631 characters) is not visible before interaction | |||||||||||||||||||
What we check How much of the page's text is hidden behind a click or an expander before anyone interacts. What good looks like The page's text is visible as it loads, not held behind a click or an expander. Why it matters Half or more of the page's text is present but never shown, so what readers see and what it contains diverge. What we found on this page 64% of the page's text (10684 of 16631 characters) is not visible before interaction The evidence, taken from the page as served How much of this page is hidden until something is clicked
Counted on the page once its scripts had run, before anything was clicked.
| |||||||||||||||||||||
| MS04 | warning | the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<button class="batcom-flyout__closebutton">Close <span class="batcom-flyout__closeicon"/> </button> | |||||||||||||||||||||
▸https://www.bat.com/who-we-are/our-corporate-governance/standards-of-business-conduct 9.6% Extremely likely 13 out of 60
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 200 | |||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 200 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="Conducting our business with honesty, integrity and transparency is not only the right thing to do, but is also critical to the continuing development of a business that is responsible, successful and"/> | |||||||||||||
| CT09 | fail | 3 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 3 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Standards of Business Conduct H2 Standards of Business Conduct H4 Conducting our business with honesty, integrity and transparency is not only the level jumped H3 What our Standards cover H4 Our Standards of Business Conduct 2026 H3 Regularly reviewed and updated H3 Anti-bribery and corruption H3 Confidential Speak Up procedures H4 Raising concerns and reporting H3 Compliance H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 2 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”, “standards of business conduct”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 2 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”, “standards of business conduct”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a itemprop="url" aria-current="page" class="cmp-navigationcorp__item-link" title="Standards of Business Conduct" href="/who-we-are/our-corporate-governance/standards-of-business-conduct"> <span class="cmp-navigationcorp__item-link-title">Standards of Business Conduct</span></a> <a data-link-type="disclosure" href="https://batsobc.com/" target="_blank">Standards of Business Conduct</a> <a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fwho-we-are%2Four-corporate-governance%2Fstandards-of-business-conduct" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… <a href="/who-we-are/our-corporate-governance/standards-of-business-conduct">Standards of Business Conduct</a> Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 3 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 3 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Standards of Business Conduct H2 Standards of Business Conduct H4 Conducting our business with honesty, integrity and transparency is not only the level jumped H3 What our Standards cover H4 Our Standards of Business Conduct 2026 H3 Regularly reviewed and updated H3 Anti-bribery and corruption H3 Confidential Speak Up procedures H4 Raising concerns and reporting H3 Compliance H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-c3ba0b29f8-cta-7f6b2b5740" data-cmp-data-layer="{"batcom-teaser-c3ba0b29f8-cta-7f6b2b5740":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||
| MS12 | fail | 1 of 1 data tables lack header cells: values are extracted without column meaning | |||||||||||
What we check Data tables mark their header cells, so each number stays tied to its column. What good looks like Data tables mark their header cells, so each value stays tied to its column. Why it matters Without headers a figure is extracted as a bare number, so it gets quoted against the wrong heading. What we found on this page 1 of 1 data tables lack header cells: values are extracted without column meaning The evidence, taken from the page as served The data tables with no header cells, as served
<table cellpadding="1" cellspacing="0" border="1"> <tbody><tr><td colspan="2"><b>Download translated PDFs</b></td> </tr><tr><td><p><a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-Arabic.pdf" target="_blank" data-link-type="download">SoBC - Arabic… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: data tables mark their header cells, so each value stays tied to its column. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it does not move the misinformation reading This failure makes content harder to reach or to read, but it is not a direct route to a wrong answer, so it is held out of the misinformation reading rather than counted twice. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| DS28 | fail | SEVERE: 32% of the links to sources on other sites did not open when they were checked. 1 points at a page that no longer exists, and 1 was refused, meaning the page is still published but the other site turns away requests that do not come from a person using a browser. That is 2 of the 5 checked, which stand for the 10 links to other sites on this page | |||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page SEVERE: 32% of the links to sources on other sites did not open when they were checked. 1 points at a page that no longer exists, and 1 was refused, meaning the page is still published but the other site turns away requests that do not come from a person using a browser. That is 2 of the 5 checked, which stand for the 10 links to other sites on this page The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| AR03 | warning | 15% of the words a person sees are missing from the page as first delivered (9626 characters delivered, 12717 once the page finishes building itself) | |||||||||||
What we check How much of the page's content exists only once JavaScript has run. What good looks like The page's content is present in the served HTML, so it does not depend on JavaScript to appear. Why it matters The page is fetched successfully but the answer is missing from what comes back, so the content is absent rather than wrong. What we found on this page 15% of the words a person sees are missing from the page as first delivered (9626 characters delivered, 12717 once the page finishes building itself) The evidence, taken from the page as served Content present only after JavaScript runs (absent from the served HTML)
The served HTML carries 9626 characters of text; after JavaScript runs the page carries 12717. 15% of the rendered page's words are absent from the HTML the server sends. What the server sends (raw HTML text, first 280 chars) British American Tobacco - Standards of Business Conduct Skip to main content Media Stories and features Press Releases Subscribe to News Alert Media contacts Social media Contact us Careers Who we are Our business Our business model At a glance Our values Our people Life at BAT …
Words present only after JavaScript runs
4570ableactiveallowalwaysanalyticsaroundbackbecausebeingbillingblockingbrowserbuttoncancelcartcentercheckboxchoicechoiceschooseclickconsentcookiescountdevicedisabledoeseasilyembedenableenhanceensuringessentialexampleexpectexperiencefilterfindingformfunctionalfunctionalityhelpshowevericonidentifyimpactimproveincludeintolabellanguagelikelistlookingmostlymovenecessarynumberoperation
| |||||||||||||
| AR11 | warning | 15% of the words a person sees are missing from the page as first delivered, which is 0.76 of its finished size | |||||||||||
What we check The measured gap between the page as served and the page after the browser has built it. What good looks like The served page and the fully built page carry the same words, with little gap between them. Why it matters Two versions of the same page exist, and which words come back depends on which version was fetched. What we found on this page 15% of the words a person sees are missing from the page as first delivered, which is 0.76 of its finished size The evidence, taken from the page as served This page as first delivered, against the page once its scripts have run
Both were taken from the same request. The first is what the site sent; the second is what a person ends up looking at.
| |||||||||||||
| AR12 | warning | 50% of the page's text (6454 of 12717 characters) is not visible before interaction | |||||||||||
What we check How much of the page's text is hidden behind a click or an expander before anyone interacts. What good looks like The page's text is visible as it loads, not held behind a click or an expander. Why it matters Half or more of the page's text is present but never shown, so what readers see and what it contains diverge. What we found on this page 50% of the page's text (6454 of 12717 characters) is not visible before interaction The evidence, taken from the page as served How much of this page is hidden until something is clicked
Counted on the page once its scripts had run, before anything was clicked.
| |||||||||||||
| MS04 | warning | the page's naming markup is faulty: 21 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 21 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-Arabic.pdf" target="_blank" data-link-type="download" rel="noopener">SoBC - Arabic</a> <a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-Bengali.pdf" target="_blank" data-link-type="download" rel="noopener">SoBC - Bengali</a> <a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-Chinese.pdf" target="_blank" data-link-type="download" rel="noopener">SoBC - Chinese</a> <a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-English.pdf" target="_blank" data-link-type="download" rel="noopener"><b>SoBC - English (Group)</b></a> <a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-English-American.pdf" target="_blank" data-link-type="download" rel="noopener">SoBC - English (Reynolds American)</a> <a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-French-Canadian.pdf" target="_blank" data-link-type="download" rel="noopener">SoBC - French (Canadian)</a> <a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-German.pdf" target="_blank" data-link-type="download" rel="noopener">SoBC - German</a> <a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-Indonesian.pdf" target="_blank" data-link-type="download" rel="noopener">SoBC - Indonesian</a> <a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-Japanese.pdf" target="_blank" data-link-type="download" rel="noopener">SoBC - Japanese</a> <a href="/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/standards-of-business-conduct/SoBC-2026-Korean.pdf" target="_blank" data-link-type="download" rel="noopener">SoBC - Korean</a> …and 11 more like it. | |||||||||||||
▸https://www.bat.com/sustainability-and-esg/our-sustainability-strategy 54.7% Likely 10 out of 60
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Our sustainability strategy H2 Our sustainability strategy H3 As we transition to A Better Tomorrow™, we are committed to doing so responsibly H3 Our Group sustainability strategy H3 Tobacco Harm Reduction H3 Climate H3 Nature H3 Circularity H3 Communities H3 2030 targets H3 Our new 2030 sustainability targets H4 CLIMATE H4 NATURE H4 CIRCULARITY H4 COMMUNITIES H4 Notes: H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fsustainability-and-esg%2Four-sustainability-strategy" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Our sustainability strategy H2 Our sustainability strategy H3 As we transition to A Better Tomorrow™, we are committed to doing so responsibly H3 Our Group sustainability strategy H3 Tobacco Harm Reduction H3 Climate H3 Nature H3 Circularity H3 Communities H3 2030 targets H3 Our new 2030 sustainability targets H4 CLIMATE H4 NATURE H4 CIRCULARITY H4 COMMUNITIES H4 Notes: H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-e3d0bb989e-cta-0b630cf0e5" data-cmp-data-layer="{"batcom-teaser-e3d0bb989e-cta-0b630cf0e5":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| AR03 | warning | 14% of the words a person sees are missing from the page as first delivered (12305 characters delivered, 15576 once the page finishes building itself) | |||||||||||
What we check How much of the page's content exists only once JavaScript has run. What good looks like The page's content is present in the served HTML, so it does not depend on JavaScript to appear. Why it matters The page is fetched successfully but the answer is missing from what comes back, so the content is absent rather than wrong. What we found on this page 14% of the words a person sees are missing from the page as first delivered (12305 characters delivered, 15576 once the page finishes building itself) The evidence, taken from the page as served Content present only after JavaScript runs (absent from the served HTML)
The served HTML carries 12305 characters of text; after JavaScript runs the page carries 15576. 14% of the rendered page's words are absent from the HTML the server sends. What the server sends (raw HTML text, first 280 chars) British American Tobacco - Our sustainability strategy Skip to main content Media Stories and features Press Releases Subscribe to News Alert Media contacts Social media Contact us Careers Who we are Our business Our business model At a glance Our values Our people Life at BAT Em…
Words present only after JavaScript runs
20264570ableacceptactiveallowalwaysanalyticsapplyaroundbackbecausebillingblockingbrowserbuttoncancelcartcentercheckboxchoicechoiceschooseclearclickconsentcookiescountdevicedisabledoeseacheasilyenableenablesenhanceensuringexampleexpectexperiencefilterfindingformfunctionalfunctionalitygivehelpshowevericonidentifyimproveincludeinterestintolabellanguagelikelistlookingmake
| |||||||||||||
| AR11 | warning | 14% of the words a person sees are missing from the page as first delivered, which is 0.79 of its finished size | |||||||||||
What we check The measured gap between the page as served and the page after the browser has built it. What good looks like The served page and the fully built page carry the same words, with little gap between them. Why it matters Two versions of the same page exist, and which words come back depends on which version was fetched. What we found on this page 14% of the words a person sees are missing from the page as first delivered, which is 0.79 of its finished size The evidence, taken from the page as served This page as first delivered, against the page once its scripts have run
Both were taken from the same request. The first is what the site sent; the second is what a person ends up looking at.
| |||||||||||||
| MS04 | warning | the page's naming markup is faulty: 6 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 6 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<a href="/strategy-and-purpose/reducing-harm" target="_self" data-link-type="internal"><b>Read about Tobacco Harm Reduction</b></a> <a href="https://www.bat.com/sustainability-and-esg/climate" target="_blank" rel="noopener"><b>Read about Climate</b></a> <a href="https://www.bat.com/sustainability-and-esg/nature" target="_blank" rel="noopener"><b>Read about Nature</b></a> <a href="https://www.bat.com/sustainability-and-esg/circularity" target="_blank" rel="noopener"><b>Read about Circularity</b></a> <a href="https://www.bat.com/sustainability-and-esg/communities" target="_blank" rel="noopener"><b>Read about Communities</b></a> <button class="batcom-flyout__closebutton">Close <span class="batcom-flyout__closeicon"/> </button> | |||||||||||||
▸https://www.bat.com/media/press-releases 53.9% Likely 11 out of 60
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 43 | |||||||||||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 43 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="British American Tobacco plc news releases."/> | |||||||||||||||||||||
| CT09 | fail | 1 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 1 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H2 Find out what's happening H3 Press releases H4 Half-Year Report for the six months to 30 June 2026 H4 Creating a stronger BAT through Fit2Win transformation programme H4 2026 First Half Pre-Close Trading Update H4 BAT launches initiative to encourage participation of consumers and retail partn H4 BAT accelerates technology transformation with new Bengaluru Hub H4 AGM 2026: Chair's address H4 BAT Appoints Chief Financial Officer H4 EU tobacco law evaluation jeopardises 2040 smoke-free goal by sidelining crucial H4 BAT to Present at CAGNY; Reaffirming 2026 Full-Year Guidance H4 Preliminary results for the year ended 31 December 2025 H4 Board Update – Chair Extension and Senior Independent Director Appointment H4 BAT recognised as a Global Top Employer for ninth consecutive year H3 Press releases H4 From Transformation to Sustained Performance: Zak Brown joins The Smokeless Word H4 Committee Change H4 2025 Full Year Pre-Close Trading Update H4 Completion of Block Trade of ITC Hotels Shares H4 Proposed Block Trade of ITC Hotels Shares H4 Velo & The McLaren Formula 1® Team Unveil Abu Dhabi GP Livery Inspired By Fans’ H4 BAT and the Channel Islands Coop roll out Yoti facial age estimation in stores t H4 Board Change H4 70% of Policy Experts Continue to Misidentify Nicotine As Primary Cause of Smoki H4 BAT Announces Management Board Changes H4 Velo & The McLaren Formula 1 Team Reveal the Ultimate Fan Experience with “Live H4 BAT announces Board Change H4 Half-Year Report for the six months to 30 June 2025 H4 BAT and Accenture announce partnership to transform global operations and supply H4 BAT Announces Management Board Changes H4 BAT Showcases Latest Science and Innovation in Tobacco Harm Reduction at Global H4 New podcast explores new perspectives on Building a Smokeless World H4 BAT launches glo™ Hilo – a new premium device featuring revolutionary TurboStart H4 2025 First Half Pre-Close Trading Update H4 BAT Reinforces Commitment to Preventing Underage Access to Tobacco and Nicotine H4 Completion of Block Trade of ITC Shares H4 Launch of Proposed Block Trade of ITC Shares H4 BAT celebrates its fifth consecutive Financial Times Climate Leader Recognition H4 2025 AGM: Chair's address H4 BAT’s corporate venture unit Btomorrow Ventures (BTV) announces new MD H4 Canada update - Court’s sanction of plan of compromise and arrangement in CCAA p H4 Preliminary results for the year ended 31 December 2024 H4 British American Tobacco (BAT) receives “Triple-A” rating from CDP for strong en H4 Board and Committee Changes H3 Press releases H4 Btomorrow Ventures launches new £200m fund to invest in wellbeing and stimulatio H4 2024 Full Year Pre-Close Trading Update H4 Canada - CCAA Update H4 BAT celebrates hosting a successful CORESTA Congress 2024 H4 BAT Capital Markets Day H4 British American Tobacco launches major global initiative to make A Smokeless Wo H4 Management Board Changes Announced H4 BAT to host CORESTA Congress 2024 - The world's leading conference on tobacco an H4 British American Tobacco - Half-Year Report for the six months to 30 June 2024 H4 Vuse Alto receives U.S. FDA Marketing Authorisation H4 2024 First Half Pre-Close Trading Update H4 2024 AGM: Chair's address H4 BAT and McLaren Racing announce multi-year partnership extension H4 Share buy-back programme H4 Completion of Block Trade of Ordinary Shares in ITC Limited H4 Launch of Proposed Block Trade of up to 436,851,457 Ordinary Shares in ITC Limit H4 BAT opens £30m Innovation Centre for New Category products in Southampton H4 Preliminary results for the year ended 31 December 2023 H4 Board and Committee Changes H4 BAT and PMI Announce Global Patent Settlement H4 BAT launches Hyper Pro, its most innovative glo device yet H3 Press releases H4 2023 Second Half Pre-Close Trading Update H4 Directorate Change H4 We need more regulation says UK’s largest vapour manufacturer H4 BAT to Strengthen Strategic Partnership with Organigram H4 BAT Appoints Chief Financial Officer H4 BAT Appoints Chief People Officer to the Management Board H4 BAT expresses disappointment at FDA Marketing Denial Order for Vuse Alto Menthol H4 Directorate Change H4 BAT outlines five-step plan towards better regulation of vapour sector H4 Russia business update – sale completed H4 Russia business update – agreement signed H4 Half-Year Report for the six months to 30 June 2023 H4 BAT announces changes to Management Board structure and composition H4 2023 First Half Pre-Close Trading Update H4 New clinical data on Vuse illustrates beneficial public health impact of Tobacco H4 BAT announces Tadeu Marroco as Chief Executive H4 BAT reaches agreement with U.S. Department of Justice (DOJ) and U.S. Department H4 2023 AGM: Chair's address H4 Marketing Denial Order issued for Vuse Solo Menthol 4.8% H4 BAT signals step change with its first combined Annual ESG Report H4 Preliminary announcement - year ended 31 December 2022 H4 BAT included in prestigious 2023 Bloomberg Gender-Equality Index H4 Driving Faster Transformation at BAT: Management Board changes and a new regiona H4 MDO for Vuse Vibe Tank Menthol 3.0% and Vuse Ciro Cartridge Menthol 1.5% H4 BAT recognised as a Global Top Employer for a sixth consecutive year H3 Press releases H4 BAT celebrates inclusion in two leading global sustainability rankings H4 Reduced-risk* potential of Velo as cigarette alternative reinforced by innovativ H4 2022 Second Half Pre-Close Trading Update H4 BAT invests in Charlotte’s Web, a leading US producer of hemp extract wellness p H4 BAT joins call for action on biodiversity H4 Innovative new study to compare the health impact of Velo modern oral nicotine p H4 BAT publishes ambitious roadmap to reach Net Zero by 2050 H4 BAT calls for collaboration to accelerate tobacco harm reduction H4 BAT announces Sanity Group investment H4 New study reveals positive long-term impact of switching from cigarettes to glo™ H4 BAT appoints Chief Sustainability Officer H4 BAT’s ambitious 2030 decarbonisation target approved by SBTi H4 Half-Year Report to 30 June 2022 H4 BAT announces Board changes H4 Watch: HYR 2022 highlights animation H4 BAT launches glo™ hyper X2: latest, state-of-the-art heated tobacco product H4 2022 First Half Pre-Close Trading Update H4 BAT conducts first real-world study assessing the impact of Vuse on indicators f H4 Vuse Vibe and Vuse Ciro receive US FDA Marketing Authorization H4 2022 AGM: Chairman's address H4 BAT notes FDA announcement on product standard for menthol cigarettes H4 BAT Announces Management Board Change H4 Russia business update and revised guidance H4 BAT statement on Ukraine/Russia H4 BAT to present at the 2022 Consumer Analyst Group of New York (CAGNY) conference H4 BAT appoints new Chief Medical Officer H4 Share buyback programme H4 Preliminary announcement - year ended 31 December 2021 H4 British American Tobacco announces Board changes H4 BAT recognised as Global Top Employer for fifth consecutive year H4 BAT launches virtual R&D Visitor Experience, showcasing the world-class science H4 BAT creates biotech company KBio to accelerate development of existing plant-bas H3 Press releases H4 BAT’s Vuse wins prestigious SEAL Sustainable Product Award H4 2021 Second Half Pre-close Trading Update H4 BAT statement on patent enforcement success in the U.S. H4 BAT in Dow Jones Sustainability Indices for 20th consecutive year - and only tob H4 BAT announces commitment to set target for net zero emissions by 2050 H4 Vuse Solo first to receive US FDA vapour marketing authorisation H4 2021 Annual General Meeting – Voting Results Update Statement H4 BAT's response to yesterday's U.S. International Trade Commission Final Determin H4 BAT emphatically rejects mischaracterisation of anti-illicit trade activities H4 BAT’s response to today’s FDA announcement H4 BAT’s Vuse becomes the number one global vaping brand H4 Half-Year Report to 30 June 2021 H4 Vuse makes waves as the first global carbon neutral vape brand H4 Building A Better Tomorrow™. Responsibly H4 New study shows completely switching to BAT’s Tobacco Heating Product, glo, has H4 2021 First Half Pre-Close Trading Update H4 BAT makes Gartner Supply Chain top 20 for the first time H4 BAT’s Vuse becomes world’s first global carbon neutral vape brand H4 World Vape Day: BAT publishes comprehensive review highlighting 10 years of scie H4 BAT recognised as a Climate Leader by the Financial Times H4 2021 AGM: Chairman’s Statement H4 British American Tobacco Notice of Board Changes and Withdrawal of AGM Resolutio H4 BAT enters strategic R&D collaboration to accelerate ‘beyond nicotine’ strategy H4 BAT Unveils Ambitious New Environmental Targets in 2020 ESG Report H4 BAT launches ESG Week to amplify sustainability progress H4 BAT accelerates its enterprise transformation towards A Better Tomorrow™ H4 Preliminary announcement - year ended 31 December 2020 H4 We strike gold again in sustainability H4 BAT recognised as Global Top Employer for fourth consecutive year H4 BAT announces succession for Director, Operations H4 SFO closes investigation into BAT H3 Press releases H4 BAT progresses COVID-19 candidate vaccine into Phase I human clinical trials H4 BAT launches tobacco industry-first Human Rights Report H4 2020 Second Half Pre-close Trading Update H4 BAT awarded A-score and named in top 200 ‘A List’ companies for climate change a H4 British American Tobacco announces Board appointment H4 BAT recognised as a Financial Times Diversity Leader for second consecutive year H4 BAT named in Dow Jones Sustainability Indices for 19th consecutive year H4 New study finds that smokers switching exclusively to glo significantly reduce t H4 BAT strengthens its US New Category portfolio: Announces acquisition of Dryft Mo H4 BAT to appoint new Chairman H4 2020 Annual General Meeting – Voting Results Update Statement H4 BAT advocates for a collaborative, stakeholder-inclusive approach to accelerate H4 BAT announces appointment of Dr Hutan Ashrafian as Chief Medical Officer H4 British American Tobacco announces Board appointment H4 BAT completes PMTA submissions for the broadest portfolio in the U.S. H4 BAT resumes sales in South Africa H4 British American Tobacco (“BAT”) announces new President for US business, Reynol H4 Half-year Report to 30 June 2020 H4 2020 First Half Pre-Close Trading Update H4 BAT's response to WTO plain packaging decision H4 BAT makes progress on COVID-19 vaccine & provides community support H4 Chairman’s AGM Statement - 2020 outlook H4 BAT working on potential COVID-19 vaccine through US bio-tech subsidiary H4 BAT launches sustainability agenda to deliver A Better Tomorrow H4 BAT Capital Markets Webcast: Building A Better Tomorrow H4 British American Tobacco signs £6 billion SONIA and SOFR linked revolving credit H4 Preliminary announcement - year ended 31 December 2019 H4 British American Tobacco named as one of the most sustainable companies in the w H4 British American Tobacco recognised as global leader for engaging its supply cha H4 Vype wins at Product of the Year awards again! H4 British American Tobacco named on CDP ‘A List’ for leading effort against climat H4 British American Tobacco welcomes FDA guidance as a further step toward a sustai H3 Press releases H4 British American Tobacco awarded leader status in Global Child Forum report H4 British American Tobacco announces enhanced partnership with McLaren Racing H4 British American Tobacco to focus on three global new category brands to further H4 VUSE Premarket Tobacco Product Application filed for substantive review by the F H4 2019 Second Half Pre-Close Trading Update H4 British American Tobacco named as a Diversity Leader by the Financial Times H4 2019 Annual General Meeting – Voting Results Update Statement H4 British American Tobacco’s U.S. subsidiary submits vapour Premarket Tobacco Appl H4 A statement from British American Tobacco on the need for robust and effective v H4 British American Tobacco is the only tobacco company featured in prestigious Dow H4 Simplifying BAT to drive New Category growth H4 British American Tobacco supports robust product quality standards, U.S. youth a H4 British American Tobacco's response to U.S. Food and Drug Administration announc H4 British American Tobacco expands its New Category portfolio with the launch of t H4 Green light for Twisp acquisition to further drive BAT’s strategic New Category H4 Half-year Report to 30 June 2019 H4 British American Tobacco announces Board appointment H4 British American Tobacco express their condolences to the family of Sir Patrick H4 First Half Pre-close Trading Update 2019 H4 British American Tobacco unveils transformed science website H4 Speech by Richard Burrows, Chairman, at the British American Tobacco p.l.c. 2019 H4 BAT’s 2018 Sustainability Report underlines Group’s commitment to Transforming T H4 Response to FDA statement on preventing youth access to flavoured products H4 Imperial Tobacco Canada Ltd. files for CCAA H4 Accounting Consequences of Quebec Judgment H4 Quebec tobacco class action Court of Appeal judgment issued H4 Quebec tobacco class action Court of Appeal judgment issued H4 British American Tobacco announces Finance Director's retirement H4 Feb 28 - Preliminary announcement - year ended 31 December 2018 H4 British American Tobacco partners globally with McLaren H4 British American Tobacco recognised again as a Global Top Employer H3 Press releases H4 British American Tobacco announces Management Board changes H4 Second Half Pre-close Trading Update 2018 H4 British American Tobacco announces Board change H4 British American Tobacco unveils new-to-world vaping technology H4 British American Tobacco launches 2018 Harm Reduction Focus Report H4 British American Tobacco named as a Top Employer in Asia Pacific H4 British American Tobacco’s response to FDA announcement H4 British American Tobacco announces new Regional Director, Americas and Sub-Sahar H4 Helping to make new parents’ lives that little bit easier: BAT pledges new benef H4 Update ahead of analyst and investor meetings H4 British American Tobacco announces Chief Marketing Officer succession H4 British American Tobacco announces CEO succession H4 British American Tobacco announces CEO's intention to retire on 1 April 2019 H4 British American Tobacco only tobacco company featured in prestigious Dow Jones H4 BAT welcomes House of Commons Science and Technology Committee e-cigarette repor H4 Half-year Report to 30 June 2018 H4 First Half Pre-close Trading Update 2018 H4 BAT to invest €800 million in its Romanian factory by 2022 to support the Europe H4 BAT’s Science and Technology Report published H4 Speech by Richard Burrows, Chairman, at the British American Tobacco p.l.c. 2018 H4 British American Tobacco p.l.c. Annual General Meeting statement H4 British American Tobacco continues its commitment to transform tobacco with glo H4 British American Tobacco underlines commitment to transforming tobacco in latest H4 Feb 22 - Preliminary announcement - year ended 31 December 2017 H4 British American Tobacco recognised as a Global Top Employer H4 Impact on British American Tobacco of new US tax legislation H3 Press releases H4 Second Half Pre-close Trading Update 2017 H4 Second Interim Dividend for the year ending 31 December 2017: South Africa Branc H4 British American Tobacco launches glo™ in Romania H4 British American Tobacco announces Management Board changes H4 British American Tobacco launches Sustainable Agriculture and Farmer Livelihoods H4 Capital Markets Day H4 Geographic expansion of glo™ continues with launch in Russia H4 Management Board Changes H4 British American Tobacco p.l.c. recommends stockholders reject mini-tender offer H4 Update on Ongoing Investigation into Misconduct Allegations H4 Half-year Report to 30 June 2017 H4 BAT Completes Acquisition of Reynolds H4 Appointment of directors in connection with acquisition of Reynolds American Inc H4 Proposed Acquisition of Reynolds American Inc. approved by BAT shareholders H4 Publication of Circular and Prospectus in connection with the Proposed Acquisiti H4 First half Pre-close Trading Update 2017 H4 BAT Korea expands its Sacheon factory to support the growth of Tobacco Heating P H4 British American Tobacco drives Tobacco Heating Product expansion with Canada la H4 Speech by Richard Burrows, Chairman, at the British American Tobacco p.l.c. 2017 H4 Proposed acquisition of Reynolds - BAT obtains antitrust approval in Japan H4 New Sustainability Report shows how BAT is responding to a changing world H4 Proposed acquisition of Reynolds - U.S. antitrust condition satisfied H4 British American Tobacco launches 2016 Harm Reduction Report H4 Feb 23 - Preliminary announcement - year ended 31 December 2016 H4 British American Tobacco recognised as a Top Employer in Europe for seventh cons H4 BAT announces agreement to acquire Reynolds H4 British American Tobacco and Kind Consumer announce a new approach to commercial H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H3 Press releases H4 British American Tobacco launches Vype-branded flagship store and introduces the H4 British American Tobacco’s statement on the English Court of Appeal’s plain pack H4 British American Tobacco launches glo™ – a new-to-world Tobacco Heating Product H4 BAT announces Board and Management Board changes H4 Interim Management Statement for the nine months ended 30 September 2016 H4 BAT proposes merger with Reynolds H4 BAT named one of the world’s most diverse and inclusive companies H4 BAT encouraged by findings of BMJ e-cigarette study H4 British American Tobacco ranked number one in good governance index H4 British American Tobacco launches 2016 Corporate Behaviour Report H4 Half-year Report to 30 June 2016 H4 British American Tobacco announces Board appointment H4 British American Tobacco’s response to the judgment in the UK plain packaging le H4 BAT’s statement on the Court of Justice of the European Union judgment on the EU H4 British American Tobacco acquires UK e-cigarette company Ten Motives H4 Speech by Richard Burrows, Chairman, at the British American Tobacco p.l.c. 2016 H4 Interim Management Statement for the three months ended 31 March 2016 H4 British American Tobacco's 2015 Sustainability Summary keeps an open door on peo H4 British American Tobacco is named as ‘Top Employer in Europe’ for sixth consecut H4 Preliminary announcement - year ended 31 December 2015 H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT10 | fail | the page carries no main heading, so nothing on it states what the page is about | |||||||||||||||||||
What we check The page carries a main heading (the H1), its own statement of what the page is about. What good looks like The page carries a single main heading stating what it is about. Why it matters With no main heading the page never states its own subject, so the sections beneath it hang from nothing. What we found on this page the page carries no main heading, so nothing on it states what the page is about The evidence, taken from the page as served The headings on this page, as served
The page was read for a top-level heading. These are the headings it does carry. <h2 class="cmp-teaser__title"> Find out what's happening </h2> <h3 class="cmp-title__text "> Press releases </h3> <h4 class="cmp-list__item-title ">Half-Year Report for the six months to 30 June 2026</h4> <h4 class="cmp-list__item-title ">Creating a stronger BAT through Fit2Win transformation programme</h4> <h4 class="cmp-list__item-title ">2026 First Half Pre-Close Trading Update</h4> <h4 class="cmp-list__item-title ">BAT launches initiative to encourage participation of consumers and retail partners in EU’s Call for Evidence </h4> <h4 class="cmp-list__item-title ">BAT accelerates technology transformation with new Bengaluru Hub</h4> <h4 class="cmp-list__item-title ">AGM 2026: Chair's address</h4> <h4 class="cmp-list__item-title ">BAT Appoints Chief Financial Officer</h4> <h4 class="cmp-list__item-title ">EU tobacco law evaluation jeopardises 2040 smoke-free goal by sidelining crucial evidence</h4> …and 313 more like it. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page carries a single main heading stating what it is about. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-ced40bf7b7" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-7d53ca0f29" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-0e96d25f94" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 5 link wording(s) are used for more than one destination, so the same words point at different pages: “1 march 2019 quebe”, “10 february 2025 b”, “12 september 2019” and 2 more; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 5 link wording(s) are used for more than one destination, so the same words point at different pages: “1 march 2019
quebe”, “10 february 2025
b”, “12 september 2019” and 2 more; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="British American Tobacco (BAT) receives “Triple-A” rating from CDP for strong environmental leadership" data-tags="bat:rss/press-releases" href="/media/press-releases/_2025/february/bat-receives-triple-a-rating-from-cdp"> <div class=… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Board and Committee Changes" data-tags="bat:rss/press-releases" href="/media/press-releases/_2025/february/board-and-committee-changes"> <div class="cmp-list__item-content"> <div class="cmp-list__item-date">10 February 2025 </div> <h… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="British American Tobacco launches major global initiative to make A Smokeless World a reality" data-tags="bat:rss/press-releases" href="/media/press-releases/_2024/september/bat-launches-major-initiative-to-make-a-smokeless-world-a-r… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Management Board Changes Announced" data-tags="bat:rss/press-releases" href="/media/press-releases/_2024/september/management-board-changes-announced"> <div class="cmp-list__item-content"> <div class="cmp-list__item-date">17 Septembe… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Simplifying BAT to drive New Category growth" data-tags="batcom-legacy:jack_bowles,batcom-legacy:reduce_management_layers,batcom-legacy:reduction_in_roles,batcom-legacy:faster_decision_making,batcom-legacy:the_elimination_of_duplicat… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="British American Tobacco supports robust product quality standards, U.S. youth access prevention measures and stands ready to make PMTA applications" data-tags="bat:news-release" href="/media/press-releases/_2019/september/sep-12---b… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Quebec tobacco class action Court of Appeal judgment issued" data-tags="bat:news-release" href="/media/press-releases/_2019/march/Mar_01-Quebec_tobacco_class_action_Court_of_Appeal_judgment_issued"> <div class="cmp-list__item-content… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Quebec tobacco class action Court of Appeal judgment issued" data-tags="bat:news-release" href="/media/press-releases/_2019/march/mar-01---quebec-tobacco-class-action-court-of-appeal-judgment-is"> <div class="cmp-list__item-content">… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Second Half Pre-close Trading Update 2017" data-tags="bat:news-release" href="/media/press-releases/_2017/december/Dec_13-Second_Half_Pre-close_Trading_Update_2017"> <div class="cmp-list__item-content"> <div class="cmp-list__item-dat… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Second Interim Dividend for the year ending 31 December 2017: South Africa Branch Register Finalisation Information" data-tags="bat:news-release" href="/media/press-releases/_2017/december/Dec_15-second_interim_dividend_for_the_year_… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the page's first heading is not its top-level heading, so the structure starts part way down; the heading structure skips a level in 1 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the page's first heading is not its top-level heading, so the structure starts part way down; the heading structure skips a level in 1 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H2 Find out what's happening level jumped H3 Press releases H4 Half-Year Report for the six months to 30 June 2026 H4 Creating a stronger BAT through Fit2Win transformation programme H4 2026 First Half Pre-Close Trading Update H4 BAT launches initiative to encourage participation of consumers and retail partn H4 BAT accelerates technology transformation with new Bengaluru Hub H4 AGM 2026: Chair's address H4 BAT Appoints Chief Financial Officer H4 EU tobacco law evaluation jeopardises 2040 smoke-free goal by sidelining crucial H4 BAT to Present at CAGNY; Reaffirming 2026 Full-Year Guidance H4 Preliminary results for the year ended 31 December 2025 H4 Board Update – Chair Extension and Senior Independent Director Appointment H4 BAT recognised as a Global Top Employer for ninth consecutive year H3 Press releases H4 From Transformation to Sustained Performance: Zak Brown joins The Smokeless Word H4 Committee Change H4 2025 Full Year Pre-Close Trading Update H4 Completion of Block Trade of ITC Hotels Shares H4 Proposed Block Trade of ITC Hotels Shares H4 Velo & The McLaren Formula 1® Team Unveil Abu Dhabi GP Livery Inspired By Fans’ H4 BAT and the Channel Islands Coop roll out Yoti facial age estimation in stores t H4 Board Change H4 70% of Policy Experts Continue to Misidentify Nicotine As Primary Cause of Smoki H4 BAT Announces Management Board Changes H4 Velo & The McLaren Formula 1 Team Reveal the Ultimate Fan Experience with “Live H4 BAT announces Board Change H4 Half-Year Report for the six months to 30 June 2025 H4 BAT and Accenture announce partnership to transform global operations and supply H4 BAT Announces Management Board Changes H4 BAT Showcases Latest Science and Innovation in Tobacco Harm Reduction at Global H4 New podcast explores new perspectives on Building a Smokeless World H4 BAT launches glo™ Hilo – a new premium device featuring revolutionary TurboStart H4 2025 First Half Pre-Close Trading Update H4 BAT Reinforces Commitment to Preventing Underage Access to Tobacco and Nicotine H4 Completion of Block Trade of ITC Shares H4 Launch of Proposed Block Trade of ITC Shares H4 BAT celebrates its fifth consecutive Financial Times Climate Leader Recognition H4 2025 AGM: Chair's address H4 BAT’s corporate venture unit Btomorrow Ventures (BTV) announces new MD H4 Canada update - Court’s sanction of plan of compromise and arrangement in CCAA p H4 Preliminary results for the year ended 31 December 2024 H4 British American Tobacco (BAT) receives “Triple-A” rating from CDP for strong en H4 Board and Committee Changes H3 Press releases H4 Btomorrow Ventures launches new £200m fund to invest in wellbeing and stimulatio H4 2024 Full Year Pre-Close Trading Update H4 Canada - CCAA Update H4 BAT celebrates hosting a successful CORESTA Congress 2024 H4 BAT Capital Markets Day H4 British American Tobacco launches major global initiative to make A Smokeless Wo H4 Management Board Changes Announced H4 BAT to host CORESTA Congress 2024 - The world's leading conference on tobacco an H4 British American Tobacco - Half-Year Report for the six months to 30 June 2024 H4 Vuse Alto receives U.S. FDA Marketing Authorisation H4 2024 First Half Pre-Close Trading Update H4 2024 AGM: Chair's address H4 BAT and McLaren Racing announce multi-year partnership extension H4 Share buy-back programme H4 Completion of Block Trade of Ordinary Shares in ITC Limited H4 Launch of Proposed Block Trade of up to 436,851,457 Ordinary Shares in ITC Limit H4 BAT opens £30m Innovation Centre for New Category products in Southampton H4 Preliminary results for the year ended 31 December 2023 H4 Board and Committee Changes H4 BAT and PMI Announce Global Patent Settlement H4 BAT launches Hyper Pro, its most innovative glo device yet H3 Press releases H4 2023 Second Half Pre-Close Trading Update H4 Directorate Change H4 We need more regulation says UK’s largest vapour manufacturer H4 BAT to Strengthen Strategic Partnership with Organigram H4 BAT Appoints Chief Financial Officer H4 BAT Appoints Chief People Officer to the Management Board H4 BAT expresses disappointment at FDA Marketing Denial Order for Vuse Alto Menthol H4 Directorate Change H4 BAT outlines five-step plan towards better regulation of vapour sector H4 Russia business update – sale completed H4 Russia business update – agreement signed H4 Half-Year Report for the six months to 30 June 2023 H4 BAT announces changes to Management Board structure and composition H4 2023 First Half Pre-Close Trading Update H4 New clinical data on Vuse illustrates beneficial public health impact of Tobacco H4 BAT announces Tadeu Marroco as Chief Executive H4 BAT reaches agreement with U.S. Department of Justice (DOJ) and U.S. Department H4 2023 AGM: Chair's address H4 Marketing Denial Order issued for Vuse Solo Menthol 4.8% H4 BAT signals step change with its first combined Annual ESG Report H4 Preliminary announcement - year ended 31 December 2022 H4 BAT included in prestigious 2023 Bloomberg Gender-Equality Index H4 Driving Faster Transformation at BAT: Management Board changes and a new regiona H4 MDO for Vuse Vibe Tank Menthol 3.0% and Vuse Ciro Cartridge Menthol 1.5% H4 BAT recognised as a Global Top Employer for a sixth consecutive year H3 Press releases H4 BAT celebrates inclusion in two leading global sustainability rankings H4 Reduced-risk* potential of Velo as cigarette alternative reinforced by innovativ H4 2022 Second Half Pre-Close Trading Update H4 BAT invests in Charlotte’s Web, a leading US producer of hemp extract wellness p H4 BAT joins call for action on biodiversity H4 Innovative new study to compare the health impact of Velo modern oral nicotine p H4 BAT publishes ambitious roadmap to reach Net Zero by 2050 H4 BAT calls for collaboration to accelerate tobacco harm reduction H4 BAT announces Sanity Group investment H4 New study reveals positive long-term impact of switching from cigarettes to glo™ H4 BAT appoints Chief Sustainability Officer H4 BAT’s ambitious 2030 decarbonisation target approved by SBTi H4 Half-Year Report to 30 June 2022 H4 BAT announces Board changes H4 Watch: HYR 2022 highlights animation H4 BAT launches glo™ hyper X2: latest, state-of-the-art heated tobacco product H4 2022 First Half Pre-Close Trading Update H4 BAT conducts first real-world study assessing the impact of Vuse on indicators f H4 Vuse Vibe and Vuse Ciro receive US FDA Marketing Authorization H4 2022 AGM: Chairman's address H4 BAT notes FDA announcement on product standard for menthol cigarettes H4 BAT Announces Management Board Change H4 Russia business update and revised guidance H4 BAT statement on Ukraine/Russia H4 BAT to present at the 2022 Consumer Analyst Group of New York (CAGNY) conference H4 BAT appoints new Chief Medical Officer H4 Share buyback programme H4 Preliminary announcement - year ended 31 December 2021 H4 British American Tobacco announces Board changes H4 BAT recognised as Global Top Employer for fifth consecutive year H4 BAT launches virtual R&D Visitor Experience, showcasing the world-class science H4 BAT creates biotech company KBio to accelerate development of existing plant-bas H3 Press releases H4 BAT’s Vuse wins prestigious SEAL Sustainable Product Award H4 2021 Second Half Pre-close Trading Update H4 BAT statement on patent enforcement success in the U.S. H4 BAT in Dow Jones Sustainability Indices for 20th consecutive year - and only tob H4 BAT announces commitment to set target for net zero emissions by 2050 H4 Vuse Solo first to receive US FDA vapour marketing authorisation H4 2021 Annual General Meeting – Voting Results Update Statement H4 BAT's response to yesterday's U.S. International Trade Commission Final Determin H4 BAT emphatically rejects mischaracterisation of anti-illicit trade activities H4 BAT’s response to today’s FDA announcement H4 BAT’s Vuse becomes the number one global vaping brand H4 Half-Year Report to 30 June 2021 H4 Vuse makes waves as the first global carbon neutral vape brand H4 Building A Better Tomorrow™. Responsibly H4 New study shows completely switching to BAT’s Tobacco Heating Product, glo, has H4 2021 First Half Pre-Close Trading Update H4 BAT makes Gartner Supply Chain top 20 for the first time H4 BAT’s Vuse becomes world’s first global carbon neutral vape brand H4 World Vape Day: BAT publishes comprehensive review highlighting 10 years of scie H4 BAT recognised as a Climate Leader by the Financial Times H4 2021 AGM: Chairman’s Statement H4 British American Tobacco Notice of Board Changes and Withdrawal of AGM Resolutio H4 BAT enters strategic R&D collaboration to accelerate ‘beyond nicotine’ strategy H4 BAT Unveils Ambitious New Environmental Targets in 2020 ESG Report H4 BAT launches ESG Week to amplify sustainability progress H4 BAT accelerates its enterprise transformation towards A Better Tomorrow™ H4 Preliminary announcement - year ended 31 December 2020 H4 We strike gold again in sustainability H4 BAT recognised as Global Top Employer for fourth consecutive year H4 BAT announces succession for Director, Operations H4 SFO closes investigation into BAT H3 Press releases H4 BAT progresses COVID-19 candidate vaccine into Phase I human clinical trials H4 BAT launches tobacco industry-first Human Rights Report H4 2020 Second Half Pre-close Trading Update H4 BAT awarded A-score and named in top 200 ‘A List’ companies for climate change a H4 British American Tobacco announces Board appointment H4 BAT recognised as a Financial Times Diversity Leader for second consecutive year H4 BAT named in Dow Jones Sustainability Indices for 19th consecutive year H4 New study finds that smokers switching exclusively to glo significantly reduce t H4 BAT strengthens its US New Category portfolio: Announces acquisition of Dryft Mo H4 BAT to appoint new Chairman H4 2020 Annual General Meeting – Voting Results Update Statement H4 BAT advocates for a collaborative, stakeholder-inclusive approach to accelerate H4 BAT announces appointment of Dr Hutan Ashrafian as Chief Medical Officer H4 British American Tobacco announces Board appointment H4 BAT completes PMTA submissions for the broadest portfolio in the U.S. H4 BAT resumes sales in South Africa H4 British American Tobacco (“BAT”) announces new President for US business, Reynol H4 Half-year Report to 30 June 2020 H4 2020 First Half Pre-Close Trading Update H4 BAT's response to WTO plain packaging decision H4 BAT makes progress on COVID-19 vaccine & provides community support H4 Chairman’s AGM Statement - 2020 outlook H4 BAT working on potential COVID-19 vaccine through US bio-tech subsidiary H4 BAT launches sustainability agenda to deliver A Better Tomorrow H4 BAT Capital Markets Webcast: Building A Better Tomorrow H4 British American Tobacco signs £6 billion SONIA and SOFR linked revolving credit H4 Preliminary announcement - year ended 31 December 2019 H4 British American Tobacco named as one of the most sustainable companies in the w H4 British American Tobacco recognised as global leader for engaging its supply cha H4 Vype wins at Product of the Year awards again! H4 British American Tobacco named on CDP ‘A List’ for leading effort against climat H4 British American Tobacco welcomes FDA guidance as a further step toward a sustai H3 Press releases H4 British American Tobacco awarded leader status in Global Child Forum report H4 British American Tobacco announces enhanced partnership with McLaren Racing H4 British American Tobacco to focus on three global new category brands to further H4 VUSE Premarket Tobacco Product Application filed for substantive review by the F H4 2019 Second Half Pre-Close Trading Update H4 British American Tobacco named as a Diversity Leader by the Financial Times H4 2019 Annual General Meeting – Voting Results Update Statement H4 British American Tobacco’s U.S. subsidiary submits vapour Premarket Tobacco Appl H4 A statement from British American Tobacco on the need for robust and effective v H4 British American Tobacco is the only tobacco company featured in prestigious Dow H4 Simplifying BAT to drive New Category growth H4 British American Tobacco supports robust product quality standards, U.S. youth a H4 British American Tobacco's response to U.S. Food and Drug Administration announc H4 British American Tobacco expands its New Category portfolio with the launch of t H4 Green light for Twisp acquisition to further drive BAT’s strategic New Category H4 Half-year Report to 30 June 2019 H4 British American Tobacco announces Board appointment H4 British American Tobacco express their condolences to the family of Sir Patrick H4 First Half Pre-close Trading Update 2019 H4 British American Tobacco unveils transformed science website H4 Speech by Richard Burrows, Chairman, at the British American Tobacco p.l.c. 2019 H4 BAT’s 2018 Sustainability Report underlines Group’s commitment to Transforming T H4 Response to FDA statement on preventing youth access to flavoured products H4 Imperial Tobacco Canada Ltd. files for CCAA H4 Accounting Consequences of Quebec Judgment H4 Quebec tobacco class action Court of Appeal judgment issued H4 Quebec tobacco class action Court of Appeal judgment issued H4 British American Tobacco announces Finance Director's retirement H4 Feb 28 - Preliminary announcement - year ended 31 December 2018 H4 British American Tobacco partners globally with McLaren H4 British American Tobacco recognised again as a Global Top Employer H3 Press releases H4 British American Tobacco announces Management Board changes H4 Second Half Pre-close Trading Update 2018 H4 British American Tobacco announces Board change H4 British American Tobacco unveils new-to-world vaping technology H4 British American Tobacco launches 2018 Harm Reduction Focus Report H4 British American Tobacco named as a Top Employer in Asia Pacific H4 British American Tobacco’s response to FDA announcement H4 British American Tobacco announces new Regional Director, Americas and Sub-Sahar H4 Helping to make new parents’ lives that little bit easier: BAT pledges new benef H4 Update ahead of analyst and investor meetings H4 British American Tobacco announces Chief Marketing Officer succession H4 British American Tobacco announces CEO succession H4 British American Tobacco announces CEO's intention to retire on 1 April 2019 H4 British American Tobacco only tobacco company featured in prestigious Dow Jones H4 BAT welcomes House of Commons Science and Technology Committee e-cigarette repor H4 Half-year Report to 30 June 2018 H4 First Half Pre-close Trading Update 2018 H4 BAT to invest €800 million in its Romanian factory by 2022 to support the Europe H4 BAT’s Science and Technology Report published H4 Speech by Richard Burrows, Chairman, at the British American Tobacco p.l.c. 2018 H4 British American Tobacco p.l.c. Annual General Meeting statement H4 British American Tobacco continues its commitment to transform tobacco with glo H4 British American Tobacco underlines commitment to transforming tobacco in latest H4 Feb 22 - Preliminary announcement - year ended 31 December 2017 H4 British American Tobacco recognised as a Global Top Employer H4 Impact on British American Tobacco of new US tax legislation H3 Press releases H4 Second Half Pre-close Trading Update 2017 H4 Second Interim Dividend for the year ending 31 December 2017: South Africa Branc H4 British American Tobacco launches glo™ in Romania H4 British American Tobacco announces Management Board changes H4 British American Tobacco launches Sustainable Agriculture and Farmer Livelihoods H4 Capital Markets Day H4 Geographic expansion of glo™ continues with launch in Russia H4 Management Board Changes H4 British American Tobacco p.l.c. recommends stockholders reject mini-tender offer H4 Update on Ongoing Investigation into Misconduct Allegations H4 Half-year Report to 30 June 2017 H4 BAT Completes Acquisition of Reynolds H4 Appointment of directors in connection with acquisition of Reynolds American Inc H4 Proposed Acquisition of Reynolds American Inc. approved by BAT shareholders H4 Publication of Circular and Prospectus in connection with the Proposed Acquisiti H4 First half Pre-close Trading Update 2017 H4 BAT Korea expands its Sacheon factory to support the growth of Tobacco Heating P H4 British American Tobacco drives Tobacco Heating Product expansion with Canada la H4 Speech by Richard Burrows, Chairman, at the British American Tobacco p.l.c. 2017 H4 Proposed acquisition of Reynolds - BAT obtains antitrust approval in Japan H4 New Sustainability Report shows how BAT is responding to a changing world H4 Proposed acquisition of Reynolds - U.S. antitrust condition satisfied H4 British American Tobacco launches 2016 Harm Reduction Report H4 Feb 23 - Preliminary announcement - year ended 31 December 2016 H4 British American Tobacco recognised as a Top Employer in Europe for seventh cons H4 BAT announces agreement to acquire Reynolds H4 British American Tobacco and Kind Consumer announce a new approach to commercial H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H4 Press Releases H3 Press releases H4 British American Tobacco launches Vype-branded flagship store and introduces the H4 British American Tobacco’s statement on the English Court of Appeal’s plain pack H4 British American Tobacco launches glo™ – a new-to-world Tobacco Heating Product H4 BAT announces Board and Management Board changes H4 Interim Management Statement for the nine months ended 30 September 2016 H4 BAT proposes merger with Reynolds H4 BAT named one of the world’s most diverse and inclusive companies H4 BAT encouraged by findings of BMJ e-cigarette study H4 British American Tobacco ranked number one in good governance index H4 British American Tobacco launches 2016 Corporate Behaviour Report H4 Half-year Report to 30 June 2016 H4 British American Tobacco announces Board appointment H4 British American Tobacco’s response to the judgment in the UK plain packaging le H4 BAT’s statement on the Court of Justice of the European Union judgment on the EU H4 British American Tobacco acquires UK e-cigarette company Ten Motives H4 Speech by Richard Burrows, Chairman, at the British American Tobacco p.l.c. 2016 H4 Interim Management Statement for the three months ended 31 March 2016 H4 British American Tobacco's 2015 Sustainability Summary keeps an open door on peo H4 British American Tobacco is named as ‘Top Employer in Europe’ for sixth consecut H4 Preliminary announcement - year ended 31 December 2015 H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-b201282326-cta-9b69ff036f" data-cmp-data-layer="{"batcom-teaser-b201282326-cta-9b69ff036f":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| AR12 | warning | 95% of the page's text (81099 of 84841 characters) is not visible before interaction | |||||||||||||||||||
What we check How much of the page's text is hidden behind a click or an expander before anyone interacts. What good looks like The page's text is visible as it loads, not held behind a click or an expander. Why it matters Half or more of the page's text is present but never shown, so what readers see and what it contains diverge. What we found on this page 95% of the page's text (81099 of 84841 characters) is not visible before interaction The evidence, taken from the page as served How much of this page is hidden until something is clicked
Counted on the page once its scripts had run, before anything was clicked.
| |||||||||||||||||||||
| MS04 | warning | the page's naming markup is faulty: 306 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 306 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<a href="/media/media-contacts">Media contacts</a> <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="From Transformation to Sustained Performance: Zak Brown joins The Smokeless Word podcast" href="/media/press-releases/_2025/december/from-transformation-to-sustained-performance-zak-brown-joins-the-smokeless-word-podcast"> <div class… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Committee Change" href="/media/press-releases/_2025/december/committee-change"> <div class="cmp-list__item-content"> <div class="cmp-list__item-date">12 December 2025 </div> <h4 class="cmp-list__item-title ">Committee Change</h4> <p … <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="2025 Full Year Pre-Close Trading Update" data-tags="bat:rss/press-releases" href="/media/press-releases/_2025/december/2025-full-year-pre-close-trading-update"> <div class="cmp-list__item-content"> <div class="cmp-list__item-date">9 … <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Completion of Block Trade of ITC Hotels Shares" data-tags="bat:rss/press-releases" href="/media/press-releases/_2025/december/completion-of-block-trade-of-itc-hotels-shares1"> <div class="cmp-list__item-content"> <div class="cmp-list… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Proposed Block Trade of ITC Hotels Shares" data-tags="bat:rss/press-releases" href="/media/press-releases/_2025/december/proposed-block-trade-of-itc-hotels-shares"> <div class="cmp-list__item-content"> <div class="cmp-list__item-date… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Velo & The McLaren Formula 1® Team Unveil Abu Dhabi GP Livery Inspired By Fans’ Favourite McLaren Memories" data-tags="bat:rss/press-releases" href="/media/press-releases/_2025/december/velo-mclaren-abu-dhabi-livery"> <div class=… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="BAT and the Channel Islands Coop roll out Yoti facial age estimation in stores to combat underage access" data-tags="bat:rss/press-releases" href="/media/press-releases/_2025/october/bat-and-the-channel-islands-coop-roll-out-yoti-fac… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="Board Change" data-tags="bat:rss/news-hub,bat:rss/press-releases" href="/media/press-releases/_2025/october/board-change"> <div class="cmp-list__item-content"> <div class="cmp-list__item-date">14 October 2025 </div> <h4 class="cmp-li… <a class="cmp-list__item-wrapper-js cmp-list__item-link cmp-list__item-wrapper" title="70% of Policy Experts Continue to Misidentify Nicotine As Primary Cause of Smoking-Related Disease" href="/media/press-releases/_2025/october/70pct-of-policy-makers-misidentify-nicotine-as-primary-cause-of-smoking-related-disease"> <… …and 296 more like it. | |||||||||||||||||||||
▸https://www.bat.com/media/media-contacts 10.2% Extremely likely 10 out of 60
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| CT13 | warning | main content is 15% of the page's text; the shared furniture around it dominates what is read out | |||||||||||||||||||
What we check How much of the page's text sits in the main content region rather than in surrounding furniture. What good looks like Most of the page's text sits in its main content region rather than in the surrounding furniture. Why it matters When the main region is a small slice of the page, what is pulled out is mostly menus and boilerplate. What we found on this page main content is 15% of the page's text; the shared furniture around it dominates what is read out The evidence, taken from the page as served How much of this page is its own content
Counted from the page as served, before any script runs. The rest is the menus, header and foot-of-page text that appear on other pages of the site as well.
| |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fmedia%2Fmedia-contacts" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none" xmlns="http:/… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | 24% of the links to sources on other sites did not open when they were checked. 2 were refused, meaning the page is still published but the other site turns away requests that do not come from a person using a browser. That is 2 of the 5 checked, which stand for the 9 links to other sites on this page | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page 24% of the links to sources on other sites did not open when they were checked. 2 were refused, meaning the page is still published but the other site turns away requests that do not come from a person using a browser. That is 2 of the 5 checked, which stand for the 9 links to other sites on this page The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| AR03 | fail | 33% of the words a person sees are missing from the page as first delivered (3964 characters delivered, 7055 once the page finishes building itself): a substantial share of the content exists only after JavaScript runs | |||||||||||||||||||
What we check How much of the page's content exists only once JavaScript has run. What good looks like The page's content is present in the served HTML, so it does not depend on JavaScript to appear. Why it matters The page is fetched successfully but the answer is missing from what comes back, so the content is absent rather than wrong. What we found on this page 33% of the words a person sees are missing from the page as first delivered (3964 characters delivered, 7055 once the page finishes building itself): a substantial share of the content exists only after JavaScript runs The evidence, taken from the page as served Content present only after JavaScript runs (absent from the served HTML)
The served HTML carries 3964 characters of text; after JavaScript runs the page carries 7055. 33% of the rendered page's words are absent from the HTML the server sends. What the server sends (raw HTML text, first 280 chars) British American Tobacco - Media contacts Skip to main content Media Stories and features Press Releases Subscribe to News Alert Media contacts Social media Contact us Careers Who we are Our business Our business model At a glance Our values Our people Life at BAT Embracing Diver…
Words present only after JavaScript runs
20264570ableaboutacceptactiveallowalsoalwaysanalyticsapplyareasaroundbackbecausebeingbillingblockingbrowserbuttoncancelcartcentercheckboxchoicechoiceschooseclearclickconsentcookiescountdevicedirectlydisabledoeseacheasilyembedenableenablesenhanceensuringessentialexampleexpectexperiencefilterfindingformfromfunctionalfunctionalitygivehavehelpshowevericonidentifyimpact
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's content is present in the served HTML, so it does not depend on JavaScript to appear. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 5 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| AR11 | warning | 33% of the words a person sees are missing from the page as first delivered, which is 0.56 of its finished size | |||||||||||||||||||
What we check The measured gap between the page as served and the page after the browser has built it. What good looks like The served page and the fully built page carry the same words, with little gap between them. Why it matters Two versions of the same page exist, and which words come back depends on which version was fetched. What we found on this page 33% of the words a person sees are missing from the page as first delivered, which is 0.56 of its finished size The evidence, taken from the page as served This page as first delivered, against the page once its scripts have run
Both were taken from the same request. The first is what the site sent; the second is what a person ends up looking at.
| |||||||||||||||||||||
| AR12 | warning | 84% of the page's text (5994 of 7055 characters) is not visible before interaction | |||||||||||||||||||
What we check How much of the page's text is hidden behind a click or an expander before anyone interacts. What good looks like The page's text is visible as it loads, not held behind a click or an expander. Why it matters Half or more of the page's text is present but never shown, so what readers see and what it contains diverge. What we found on this page 84% of the page's text (5994 of 7055 characters) is not visible before interaction The evidence, taken from the page as served How much of this page is hidden until something is clicked
Counted on the page once its scripts had run, before anything was clicked.
| |||||||||||||||||||||
| MS04 | warning | the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out | |||||||||||||||||||
What we check Accessibility markup faults: invented roles, labels pointing at nothing, live controls marked hidden. What good looks like The page's accessibility markup is sound, with valid roles and no labels pointing at nothing. Why it matters The page's own description of its controls is wrong, so what it offers and what it declares do not match. What we found on this page the page's naming markup is faulty: 1 working control(s) sit inside a section the code marks as hidden, so they work on screen but are absent from what is read out The evidence, taken from the page as served The faulty markup, as the page serves it once loaded
<button class="batcom-flyout__closebutton">Close <span class="batcom-flyout__closeicon"/> </button> | |||||||||||||||||||||
| CT34 | fail | the page's own words are thin: 579 of its 601 words are shared navigation and footer, so it reads as 96% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 579 of its 601 words are shared navigation and footer, so it reads as 96% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/contact-us 30.5% Very likely 7 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 42 | |||||||||||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 42 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="We are listening, find ways to contact us."/> | |||||||||||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Contact us H2 (no text) H4 What works best for you? level jumped H4 Leave us a message H4 Call us or find us H4 Connect to us H4 Speak up H3 Leave us a message H3 Consumers H3 Suppliers H3 Acquisition of Reynolds American Inc. – Investor FAQs H3 Fraud and scam alert H4 Recruitment fraud H4 Prize award and grant frauds H4 Other business-related fraud H4 What to do H3 You might also be interested in H2 Allyship and creating an inclusive culture H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-633250c70d" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-7c62ff11c0" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7162a2146a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Contact us H2 (no text) H4 What works best for you? level jumped H4 Leave us a message H4 Call us or find us H4 Connect to us H4 Speak up H3 Leave us a message H3 Consumers H3 Suppliers H3 Acquisition of Reynolds American Inc. – Investor FAQs H3 Fraud and scam alert H4 Recruitment fraud H4 Prize award and grant frauds H4 Other business-related fraud H4 What to do H3 You might also be interested in H2 Allyship and creating an inclusive culture H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-64dcc72930-cta-7e0bfc9f72" data-cmp-data-layer="{"batcom-teaser-64dcc72930-cta-7e0bfc9f72":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/me… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | SEVERE: 32% of the links to sources on other sites did not open when they were checked. 1 points at a page that no longer exists, and 1 was refused, meaning the page is still published but the other site turns away requests that do not come from a person using a browser. That is 2 of the 5 checked, which stand for the 9 links to other sites on this page | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page SEVERE: 32% of the links to sources on other sites did not open when they were checked. 1 points at a page that no longer exists, and 1 was refused, meaning the page is still published but the other site turns away requests that do not come from a person using a browser. That is 2 of the 5 checked, which stand for the 9 links to other sites on this page The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/investors-and-reporting/reporting/sustainability-reporting/modern-slavery-statement 32.2% Very likely 9 out of 45
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 43 | |||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 43 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="Modern slavery statement Modern slavery Act"/> | |||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Modern Slavery Statement H2 Modern Slavery Statement H3 Modern Slavery Statement 2025 H3 Previous Modern Slavery Statements H3 (no text) H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Finvestors-and-reporting%2Freporting%2Fsustainability-reporting%2Fmodern-slavery-statement" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg w… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Modern Slavery Statement H2 Modern Slavery Statement H3 Modern Slavery Statement 2025 H3 Previous Modern Slavery Statements H3 (no text) H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-a93344d8b0-cta-d3dad6a9e2" data-cmp-data-layer="{"batcom-teaser-a93344d8b0-cta-d3dad6a9e2":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT34 | fail | the page's own words are thin: 588 of its 627 words are shared navigation and footer, so it reads as 96% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 588 of its 627 words are shared navigation and footer, so it reads as 96% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
▸https://www.bat.com/media/news-and-stories 53.3% Likely 7 out of 43
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H2 Features, films and facts about BAT H4 Video: Chief Executive's take on Half-Year Results 2026 level jumped H4 The Smokeless Word: The Rewind - A Look Back at The Smokeless Word So Far H4 BAT recognised with triple win at the 2026 Red Dot Awards H4 The Smokeless Word: Live from the Lab - Elaine Round H4 Rural Development Day H4 BAT at Work Episode 2: Fail. Learn. Grow. H4 BAT wins ‘GEP Procurement Team Award’ at World Procurement Awards H4 The Smokeless Word - Damian Willemse H4 The Smokeless Word: Live from the Lab - Joe Gitchell H4 World Environment Day 2026: Recognising the role agriculture can play in buildin H4 BAT named one of Europe’s Best Employers by the Financial Times and Statista H4 The Smokeless Word – Jack Rankin H4 The Smokeless Word – Christian Sandström H4 Introducing BAT at Work: a new podcast sharing real stories from our people H4 The Smokeless Word – Pascale Meulemeester H4 World Earth Day 2026: Supporting the Transition to a Lower-Carbon Future H4 The Smokeless Word – Donato Raponi H4 The Smokeless Word – Tony Abbott H4 BAT wins Digital Transformation Award at Verdantix Transform Summit H4 The Smokeless Word – Boris Johnson H4 The Smokeless Word – Will Greenwood H4 The Smokeless Word – Mauricio Cantisani H4 Video: Chief Executive’s take on Preliminary Results 2025 H4 The Smokeless Word – Gavin Esler H4 The Smokeless Word – Mike Scudamore H4 BAT Achieves Prestigious CDP ‘Triple A’ Score for Environmental Leadership H4 The Smokeless Word – Nike Örbrink H4 The Smokeless Word – Derek Yach H4 The Smokeless Word – Australia's Illicit Trade Crisis H4 Omni™ Recognised for Global Award H4 Sweden’s Smoke-Free Success: What If “Big Tobacco” Is Right? H4 The Smokeless Word – Podcast Episode 6 now available H4 The Smokeless Word – Podcast Episode 5 now available H4 The Smokeless Word – Podcast Episode 4 now available H4 Video: Chief Executive's and Chief Financial Officer's take on Half-Year Results H4 The Smokeless Word – Podcast Episode 3 now available H4 BAT recognised as one of Europe’s Best Employers 2025 by the Financial Times and H4 Video: Chief Executive’s and Chief Financial Officer's take on Preliminary Resul H4 BAT recognised as a Global Top Employer for eighth consecutive year H4 BAT publishes new industry-leading commitments to responsible vaping products H4 BAT presents Omni™ at Global Tobacco and Nicotine Forum to advance Tobacco Harm H4 BAT launches new global initiative for a Smokeless World H4 Video: Chief Executive’s take on our Half-Year Results 2024 H4 BAT recognised as a Financial Times Climate Leader H4 New research further reinforces reduced-risk potential of oral nicotine pouches H4 BAT and McLaren Racing announce partnership extension H4 New Innovation Centre in Southampton H4 Animation: our 2023 Preliminary Results highlights H4 Video: Chief Executive’s take on Preliminary Results 2023 H4 BAT signs up as an Early Adopter of the TNFD H4 BAT is one of just 15 companies to be recognised as a Global Top Employer H4 A decade of portfolio transformation at BAT H4 Watch: Vuse Longitudinal Study Interim Results Overview H4 A smoke-free UK requires smarter vaping regulation H4 New investment supports the development of bioplastics H4 Diversity and inclusion heroes H4 BAT calls for smarter regulations to achieve smokefree ambitions H4 The science behind vapour products: new podcast on ResearchPod published H4 Video: Chief Executive’s take on HYR 2023 H4 Animation: our 2023 Half-Year Results highlights H4 New Innovation Hub opens in Italy H4 Solutions to plastic pollution – the BAT perspective H4 World Vape Day: New scientific study on Vuse underscores contribution vapour pro H4 Championing diversity and inclusion H4 BAT named a Financial Times 2023 climate change leader H4 BAT on Gartner: Leveraging disruption to accelerate sustainability H4 Our leading climate change engagement H4 Btomorrow Ventures names new MD H4 Velo front and centre at McLaren F1 car launch H4 Innovating for a sustainable future: Celebrating BAT’s Women in Science H4 Animation: our 2022 results highlights H4 Vuse ‘drives for change’ again with McLaren car livery H4 Focus on our Low-Carbon Transition Plan & climate targets H4 Announcing the 2022 Battle of Minds winners H4 More recognition for BAT’s Global Business Solutions H4 BAT calls for collaboration to accelerate tobacco harm reduction H4 Allyship and creating an inclusive culture H4 Welcoming Véronique Laury to our Main Board H4 Recognition for disability support H4 ‘The transformation is real, and we are delivering tangible change’ H4 A Better Tomorrow™ around the world H4 The pivotal role of suppliers in our transformation H4 2022 Half-Year Results: highlights animation H4 How we’re becoming a multi-category Enterprise of the Future H4 ‘Our transformation continues at pace’ H4 The importance of Pride in 2022 H4 Showing our solidarity with Pride H4 Innovation Hub reaches milestone H4 Our journey to a greener tomorrow H4 Let the Battle of Minds begin! H4 BTV’s Lisa has the power! H4 BAT Bangladesh makes history with ESG Report H4 Strength through diversity and inclusion H4 BAT recognised as an FT Climate Leader again H4 Vuse strikes more sustainability gold H4 Vaping facts campaign under way in Canada H4 The next stage of our transformation H4 BTV’s Lukasz named as a rising star H4 Double win at Product of the Year awards H4 2022 McLaren Racing F1 car unveiled H4 ‘It’s a very exciting time to be a scientist at BAT!’ H4 Watch: 2021 results highlights animation H4 BAT strikes gold again in sustainability H4 Showcasing our world-class science H4 Celebrating 20 years in the DJSI H4 Vuse and McLaren Racing mark world first H4 Digital award highlights our A Better Tomorrow™ transformation H4 It’s TIME to talk automation H4 Battle of Minds: and the winners are… H4 BAT lands industry-first diversity recognition H4 The importance of effective tobacco harm reduction strategy H4 National Inclusion week: Jay’s story H4 A Better Tomorrow™ Innovation Hub to launch in Italy H4 Refreshed science site showcases our research H4 2021 is a critical year for tobacco harm reduction H4 BAT in Top 20 Most Admired Shared Services Organisations (SSOs) for third year r H4 BAT wins 2nd runner-up for best ESG Report globally H4 BTV’s Lisa ranked in top 100 corporate venturing ‘powerlist’ H4 Accelerating our progress towards a sustainable future H4 Animation: Highlights of BAT’s 2021 Half-Year Results H4 Building A Better Tomorrow™. Responsibly H4 New-look careers site has ‘candidate at its heart’ H4 Vuse strikes gold for ‘Best Brand Experience’ H4 Grad Academy goes virtual again H4 Rainbow display shows we’re proud to B United H4 Our Science Innovation Report – available now H4 Btomorrow Ventures website goes live H4 Reimagine, Recreate, Restore – celebrating World Environment Day H4 We’ll still B United, no matter what H4 Biodiversity Day 2021: We’re part of the solution H4 Marina ‘honoured’ by CIO award nomination H4 ESG front and centre as Battle of Minds returns H4 Marking the World Day for Safety and Health at Work H4 BAT recognised as an ESG leader for workforce transparency H4 Award acknowledges our digital transformation H4 RD student scheme honoured again H4 Video: BAT’s 2020 full-year results H4 FYR 2020: our transformation story in focus on CNBC H4 We strike gold again in sustainability H4 Campaign highlights how BAT is building A Better Tomorrow™ H4 Celebrating our women in Science H4 Reynolds again named as a U.S. ‘best place to work for LGBTQ+ equality’ H4 Multiple employer recognition awards H4 BAT launches first CBD vaping product H4 Our COVID-19 candidate vaccine given approval to progress into human trials H4 BAT launches tobacco industry-first Human Rights Report H4 We're listed in DJSI for 19th consecutive year H4 BAT named as FT Diversity Leader again H4 How digital is driving our transformation H4 Virtual Grad Academy is a great success H4 Looking after mental health in a pandemic H4 Our CMO’s keynote speech at GTNF 2020 H4 BAT in Top 20 Most Admired Shared Services Organisations H4 Proud to B United at our R&D centre H4 Women in Science video lands 'dotCOMM' honour H4 Sustainability report wins ‘Openness and Honesty’ award H4 Event shows we're #StrongerTogether @BAT H4 Battle of Minds: our global internship case competition H4 Continuing to B United in a pandemic H4 Women in STEM: a key part of our transformation H4 WeQual recognition for Rafaela highlights importance of diversity H4 BAT named in The Gartner Supply Chain Top 25 for 2020 H4 COVID-19 vaccine: update on our progress H4 Our response to the COVID-19 pandemic H4 Building A Better Tomorrow H4 International Women's Day at BAT H4 BAT responds to latest PHE vaping report H4 Celebrating International Mother Language Day H4 Celebrating our Women in Science H4 RAI Group LGBTQ equality honour’ H4 Vype wins at Product of the Year awards again! H4 Attracting the best – Augustow style! H4 Simplifying our New Category brands H4 Our New Category ambitions in focus H4 BAT hailed for IWD best practice H4 Public Health England issues advice on vaping H4 Mental health focus: Jo’s journey H4 Looking back on an incredible summer for B United H4 U.S. Food and Drug Administration vaping illness update H4 Our business transformation in focus on CNN H4 WOW factor heralds ‘new beginning’ for women at CTC H4 10 years of our Probaho clean water initiative H4 Introducing our Women in STEM initiative H4 Our science website has been transformed H4 Transforming Tobacco in focus on CNBC H4 Pride month focus: Natalia’s story H4 Proud to B United – introducing BAT’s LGBT+ network H4 Celebrating Global Day of Parents H4 What makes a BAT person? H4 Latest Grad Academy is a huge success H4 Welcoming our new Chief Executive H4 Improving access to clean water in Fiji H4 Our global partners McLaren unveil new F1 car H4 Celebrating BAT’s female scientists H4 Assessing the reduced-risk potential of our products H4 We’re a Global Top Employer again! H4 BAT Kenya celebrates workplace transformation H4 We’re a top employer again in Asia-Pacific! H4 #NextGenBAT: our latest global graduates H4 Introducing Parents@BAT H4 Fortune favours RAI in best workplaces list H4 We’re a top employer in Africa H4 On the BAT career journey in Northern Europe H4 Learning with Lynda.com H4 Tobacco harm reduction in focus H4 BAT is recognised as a Global Top Employer H4 What our employees had to say about working at BAT H4 We are BAT - Bangladesh H4 NGP progress showcased at Capital Markets Day H4 RAI among employee engagement ‘elite’ H4 We are BAT - South Africa H4 BAT brings light to 3,000 lives in Pakistan H4 Opening the doors to a flagship store H4 Glad grads H4 Your 90-second guide to BAT H4 Inspiring women in Bangladesh H4 We are BAT – our company video H3 You might also be interested in H2 Video: 2025 Half-Year Results Highlights H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT10 | fail | the page carries no main heading, so nothing on it states what the page is about | |||||||||||||||||||
What we check The page carries a main heading (the H1), its own statement of what the page is about. What good looks like The page carries a single main heading stating what it is about. Why it matters With no main heading the page never states its own subject, so the sections beneath it hang from nothing. What we found on this page the page carries no main heading, so nothing on it states what the page is about The evidence, taken from the page as served The headings on this page, as served
The page was read for a top-level heading. These are the headings it does carry. <h2 class="cmp-teaser__title"> Features, films and facts about BAT </h2> <h4 class="cmp-list__item-title ">Video: Chief Executive's take on Half-Year Results 2026</h4> <h4 class="cmp-list__item-title ">The Smokeless Word: The Rewind - A Look Back at The Smokeless Word So Far</h4> <h4 class="cmp-list__item-title ">BAT recognised with triple win at the 2026 Red Dot Awards</h4> <h4 class="cmp-list__item-title ">The Smokeless Word: Live from the Lab - Elaine Round</h4> <h4 class="cmp-list__item-title ">Rural Development Day</h4> <h4 class="cmp-list__item-title ">BAT at Work Episode 2: Fail. Learn. Grow.</h4> <h4 class="cmp-list__item-title ">BAT wins ‘GEP Procurement Team Award’ at World Procurement Awards</h4> <h4 class="cmp-list__item-title ">The Smokeless Word - Damian Willemse</h4> <h4 class="cmp-list__item-title ">The Smokeless Word: Live from the Lab - Joe Gitchell</h4> …and 222 more like it. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page carries a single main heading stating what it is about. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-ced40bf7b7" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-7d53ca0f29" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-0e96d25f94" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 2 link wording(s) are used for more than one destination, so the same words point at different pages: “celebrating our women in science”, “read more >”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 2 link wording(s) are used for more than one destination, so the same words point at different pages: “celebrating our women in science”, “read more >”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-list__item-link cmp-list__item-link--bottom" title="Video: Chief Executive's take on Half-Year Results 2026" href="/media/news-and-stories/_2026/july/video-chief-executives-take-on-half-year-results-2026"> Read more > </a> <a class="cmp-list__item-link cmp-list__item-link--bottom" title="The Smokeless Word: The Rewind - A Look Back at The Smokeless Word So Far" href="/media/news-and-stories/_2026/july/the-smokeless-word-rewind-july-2026"> Read more > </a> <a class="cmp-list__item-link cmp-list__item-link--bottom" title="BAT recognised with triple win at the 2026 Red Dot Awards" href="/media/news-and-stories/_2026/july/bat-recognised-with-triple-win-at-2026-red-dot-awards"> Read more > </a> <a class="cmp-list__item-link cmp-list__item-link--bottom" title="The Smokeless Word: Live from the Lab - Elaine Round" href="/media/news-and-stories/_2026/july/the-smokeless-word-elaine-round"> Read more > </a> <a class="cmp-list__item-link cmp-list__item-link--bottom" title="Rural Development Day" href="/media/news-and-stories/_2026/july/rural-development-day-2026"> Read more > </a> <a class="cmp-list__item-link cmp-list__item-link--bottom" title="BAT at Work Episode 2: Fail. Learn. Grow." href="/media/news-and-stories/_2026/july/bat-at-work-episode-2-fail-learn-grow"> Read more > </a> <a class="cmp-list__item-link cmp-list__item-link--bottom" title="BAT wins ‘GEP Procurement Team Award’ at World Procurement Awards" href="/media/news-and-stories/_2026/june/bat-wins-GEP-procurement-team-award"> Read more > </a> <a class="cmp-list__item-link cmp-list__item-link--bottom" title="The Smokeless Word - Damian Willemse" href="/media/news-and-stories/_2026/june/the-smokeless-word-damian-willemse"> Read more > </a> <a class="cmp-list__item-link cmp-list__item-link--bottom" title="The Smokeless Word: Live from the Lab - Joe Gitchell" href="/media/news-and-stories/_2026/june/the-smokeless-word-joe-gitchell"> Read more > </a> <a class="cmp-list__item-link cmp-list__item-link--bottom" title="World Environment Day 2026: Recognising the role agriculture can play in building a more resilient future" href="/media/news-and-stories/_2026/june/world-environment-day-2026"> Read more > </a> Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the page's first heading is not its top-level heading, so the structure starts part way down; the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the page's first heading is not its top-level heading, so the structure starts part way down; the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H2 Features, films and facts about BAT level jumped H4 Video: Chief Executive's take on Half-Year Results 2026 level jumped H4 The Smokeless Word: The Rewind - A Look Back at The Smokeless Word So Far H4 BAT recognised with triple win at the 2026 Red Dot Awards H4 The Smokeless Word: Live from the Lab - Elaine Round H4 Rural Development Day H4 BAT at Work Episode 2: Fail. Learn. Grow. H4 BAT wins ‘GEP Procurement Team Award’ at World Procurement Awards H4 The Smokeless Word - Damian Willemse H4 The Smokeless Word: Live from the Lab - Joe Gitchell H4 World Environment Day 2026: Recognising the role agriculture can play in buildin H4 BAT named one of Europe’s Best Employers by the Financial Times and Statista H4 The Smokeless Word – Jack Rankin H4 The Smokeless Word – Christian Sandström H4 Introducing BAT at Work: a new podcast sharing real stories from our people H4 The Smokeless Word – Pascale Meulemeester H4 World Earth Day 2026: Supporting the Transition to a Lower-Carbon Future H4 The Smokeless Word – Donato Raponi H4 The Smokeless Word – Tony Abbott H4 BAT wins Digital Transformation Award at Verdantix Transform Summit H4 The Smokeless Word – Boris Johnson H4 The Smokeless Word – Will Greenwood H4 The Smokeless Word – Mauricio Cantisani H4 Video: Chief Executive’s take on Preliminary Results 2025 H4 The Smokeless Word – Gavin Esler H4 The Smokeless Word – Mike Scudamore H4 BAT Achieves Prestigious CDP ‘Triple A’ Score for Environmental Leadership H4 The Smokeless Word – Nike Örbrink H4 The Smokeless Word – Derek Yach H4 The Smokeless Word – Australia's Illicit Trade Crisis H4 Omni™ Recognised for Global Award H4 Sweden’s Smoke-Free Success: What If “Big Tobacco” Is Right? H4 The Smokeless Word – Podcast Episode 6 now available H4 The Smokeless Word – Podcast Episode 5 now available H4 The Smokeless Word – Podcast Episode 4 now available H4 Video: Chief Executive's and Chief Financial Officer's take on Half-Year Results H4 The Smokeless Word – Podcast Episode 3 now available H4 BAT recognised as one of Europe’s Best Employers 2025 by the Financial Times and H4 Video: Chief Executive’s and Chief Financial Officer's take on Preliminary Resul H4 BAT recognised as a Global Top Employer for eighth consecutive year H4 BAT publishes new industry-leading commitments to responsible vaping products H4 BAT presents Omni™ at Global Tobacco and Nicotine Forum to advance Tobacco Harm H4 BAT launches new global initiative for a Smokeless World H4 Video: Chief Executive’s take on our Half-Year Results 2024 H4 BAT recognised as a Financial Times Climate Leader H4 New research further reinforces reduced-risk potential of oral nicotine pouches H4 BAT and McLaren Racing announce partnership extension H4 New Innovation Centre in Southampton H4 Animation: our 2023 Preliminary Results highlights H4 Video: Chief Executive’s take on Preliminary Results 2023 H4 BAT signs up as an Early Adopter of the TNFD H4 BAT is one of just 15 companies to be recognised as a Global Top Employer H4 A decade of portfolio transformation at BAT H4 Watch: Vuse Longitudinal Study Interim Results Overview H4 A smoke-free UK requires smarter vaping regulation H4 New investment supports the development of bioplastics H4 Diversity and inclusion heroes H4 BAT calls for smarter regulations to achieve smokefree ambitions H4 The science behind vapour products: new podcast on ResearchPod published H4 Video: Chief Executive’s take on HYR 2023 H4 Animation: our 2023 Half-Year Results highlights H4 New Innovation Hub opens in Italy H4 Solutions to plastic pollution – the BAT perspective H4 World Vape Day: New scientific study on Vuse underscores contribution vapour pro H4 Championing diversity and inclusion H4 BAT named a Financial Times 2023 climate change leader H4 BAT on Gartner: Leveraging disruption to accelerate sustainability H4 Our leading climate change engagement H4 Btomorrow Ventures names new MD H4 Velo front and centre at McLaren F1 car launch H4 Innovating for a sustainable future: Celebrating BAT’s Women in Science H4 Animation: our 2022 results highlights H4 Vuse ‘drives for change’ again with McLaren car livery H4 Focus on our Low-Carbon Transition Plan & climate targets H4 Announcing the 2022 Battle of Minds winners H4 More recognition for BAT’s Global Business Solutions H4 BAT calls for collaboration to accelerate tobacco harm reduction H4 Allyship and creating an inclusive culture H4 Welcoming Véronique Laury to our Main Board H4 Recognition for disability support H4 ‘The transformation is real, and we are delivering tangible change’ H4 A Better Tomorrow™ around the world H4 The pivotal role of suppliers in our transformation H4 2022 Half-Year Results: highlights animation H4 How we’re becoming a multi-category Enterprise of the Future H4 ‘Our transformation continues at pace’ H4 The importance of Pride in 2022 H4 Showing our solidarity with Pride H4 Innovation Hub reaches milestone H4 Our journey to a greener tomorrow H4 Let the Battle of Minds begin! H4 BTV’s Lisa has the power! H4 BAT Bangladesh makes history with ESG Report H4 Strength through diversity and inclusion H4 BAT recognised as an FT Climate Leader again H4 Vuse strikes more sustainability gold H4 Vaping facts campaign under way in Canada H4 The next stage of our transformation H4 BTV’s Lukasz named as a rising star H4 Double win at Product of the Year awards H4 2022 McLaren Racing F1 car unveiled H4 ‘It’s a very exciting time to be a scientist at BAT!’ H4 Watch: 2021 results highlights animation H4 BAT strikes gold again in sustainability H4 Showcasing our world-class science H4 Celebrating 20 years in the DJSI H4 Vuse and McLaren Racing mark world first H4 Digital award highlights our A Better Tomorrow™ transformation H4 It’s TIME to talk automation H4 Battle of Minds: and the winners are… H4 BAT lands industry-first diversity recognition H4 The importance of effective tobacco harm reduction strategy H4 National Inclusion week: Jay’s story H4 A Better Tomorrow™ Innovation Hub to launch in Italy H4 Refreshed science site showcases our research H4 2021 is a critical year for tobacco harm reduction H4 BAT in Top 20 Most Admired Shared Services Organisations (SSOs) for third year r H4 BAT wins 2nd runner-up for best ESG Report globally H4 BTV’s Lisa ranked in top 100 corporate venturing ‘powerlist’ H4 Accelerating our progress towards a sustainable future H4 Animation: Highlights of BAT’s 2021 Half-Year Results H4 Building A Better Tomorrow™. Responsibly H4 New-look careers site has ‘candidate at its heart’ H4 Vuse strikes gold for ‘Best Brand Experience’ H4 Grad Academy goes virtual again H4 Rainbow display shows we’re proud to B United H4 Our Science Innovation Report – available now H4 Btomorrow Ventures website goes live H4 Reimagine, Recreate, Restore – celebrating World Environment Day H4 We’ll still B United, no matter what H4 Biodiversity Day 2021: We’re part of the solution H4 Marina ‘honoured’ by CIO award nomination H4 ESG front and centre as Battle of Minds returns H4 Marking the World Day for Safety and Health at Work H4 BAT recognised as an ESG leader for workforce transparency H4 Award acknowledges our digital transformation H4 RD student scheme honoured again H4 Video: BAT’s 2020 full-year results H4 FYR 2020: our transformation story in focus on CNBC H4 We strike gold again in sustainability H4 Campaign highlights how BAT is building A Better Tomorrow™ H4 Celebrating our women in Science H4 Reynolds again named as a U.S. ‘best place to work for LGBTQ+ equality’ H4 Multiple employer recognition awards H4 BAT launches first CBD vaping product H4 Our COVID-19 candidate vaccine given approval to progress into human trials H4 BAT launches tobacco industry-first Human Rights Report H4 We're listed in DJSI for 19th consecutive year H4 BAT named as FT Diversity Leader again H4 How digital is driving our transformation H4 Virtual Grad Academy is a great success H4 Looking after mental health in a pandemic H4 Our CMO’s keynote speech at GTNF 2020 H4 BAT in Top 20 Most Admired Shared Services Organisations H4 Proud to B United at our R&D centre H4 Women in Science video lands 'dotCOMM' honour H4 Sustainability report wins ‘Openness and Honesty’ award H4 Event shows we're #StrongerTogether @BAT H4 Battle of Minds: our global internship case competition H4 Continuing to B United in a pandemic H4 Women in STEM: a key part of our transformation H4 WeQual recognition for Rafaela highlights importance of diversity H4 BAT named in The Gartner Supply Chain Top 25 for 2020 H4 COVID-19 vaccine: update on our progress H4 Our response to the COVID-19 pandemic H4 Building A Better Tomorrow H4 International Women's Day at BAT H4 BAT responds to latest PHE vaping report H4 Celebrating International Mother Language Day H4 Celebrating our Women in Science H4 RAI Group LGBTQ equality honour’ H4 Vype wins at Product of the Year awards again! H4 Attracting the best – Augustow style! H4 Simplifying our New Category brands H4 Our New Category ambitions in focus H4 BAT hailed for IWD best practice H4 Public Health England issues advice on vaping H4 Mental health focus: Jo’s journey H4 Looking back on an incredible summer for B United H4 U.S. Food and Drug Administration vaping illness update H4 Our business transformation in focus on CNN H4 WOW factor heralds ‘new beginning’ for women at CTC H4 10 years of our Probaho clean water initiative H4 Introducing our Women in STEM initiative H4 Our science website has been transformed H4 Transforming Tobacco in focus on CNBC H4 Pride month focus: Natalia’s story H4 Proud to B United – introducing BAT’s LGBT+ network H4 Celebrating Global Day of Parents H4 What makes a BAT person? H4 Latest Grad Academy is a huge success H4 Welcoming our new Chief Executive H4 Improving access to clean water in Fiji H4 Our global partners McLaren unveil new F1 car H4 Celebrating BAT’s female scientists H4 Assessing the reduced-risk potential of our products H4 We’re a Global Top Employer again! H4 BAT Kenya celebrates workplace transformation H4 We’re a top employer again in Asia-Pacific! H4 #NextGenBAT: our latest global graduates H4 Introducing Parents@BAT H4 Fortune favours RAI in best workplaces list H4 We’re a top employer in Africa H4 On the BAT career journey in Northern Europe H4 Learning with Lynda.com H4 Tobacco harm reduction in focus H4 BAT is recognised as a Global Top Employer H4 What our employees had to say about working at BAT H4 We are BAT - Bangladesh H4 NGP progress showcased at Capital Markets Day H4 RAI among employee engagement ‘elite’ H4 We are BAT - South Africa H4 BAT brings light to 3,000 lives in Pakistan H4 Opening the doors to a flagship store H4 Glad grads H4 Your 90-second guide to BAT H4 Inspiring women in Bangladesh H4 We are BAT – our company video H3 You might also be interested in H2 Video: 2025 Half-Year Results Highlights H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/media/social-media 10.2% Extremely likely 9 out of 43
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT31 | warning | the page carries no description of itself in its head | |||||||||||||||||||
What we check The page provides a meta description, the short summary offered alongside its title. What good looks like The page provides a meta description. Why it matters Absent, the page gives up its one documented summary field, though Google rewrites these freely anyway. What we found on this page the page carries no description of itself in its head The evidence, taken from the page as served The head of this page, as served
The head of the page was read for a description tag. <head> <script defer="defer" type="text/javascript" src="https://rum.hlx.page/.rum/@adobe/helix-rum-js@%5E2/dist/rum-standalone.js" data-routing="env=prod,tier=publish,ams=British American Tobacco"/> <script src="https://cdn.cookielaw.org/scripttemplates/otSDKStub.js" async="" data-domain-script="5fc350fc-73ff-46f2-9463-37d2cf7f5f1a" charset="UTF-8"/> <script> (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push ({'gtm.start': new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0], j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src= 'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f); })(window,document,'script','dataLayer','GTM-TMH7W3WX');</script> <meta charset="UTF-8"/> <title> British American Tobacco - Social media </title> <meta name="template" content="evergreenpage-template"/> <meta name="viewport" content="width=device-width, initial-scale=1"/> <meta name="rendered" content="2026-07-22 13:48 +0000"/> <link rel="canonical" href="https://www.bat.com/media/social-media"/> <script type="application/ld+json"> { "@context": "https://schema.org", "@type": "WebPage", "@id": "https://www.bat.com/media/social-media/#webpa… | |||||||||||||||||||||
| CT09 | fail | 1 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 1 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Social media H2 Social media H4 X (Twitter) level jumped H4 LinkedIn H4 Facebook H4 Instagram H4 Glassdoor H4 YouTube H4 Follow us H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| CT13 | fail | main content is 14% of the page's text; the shared furniture around it dominates what is read out | |||||||||||||||||||
What we check How much of the page's text sits in the main content region rather than in surrounding furniture. What good looks like Most of the page's text sits in its main content region rather than in the surrounding furniture. Why it matters When the main region is a small slice of the page, what is pulled out is mostly menus and boilerplate. What we found on this page main content is 14% of the page's text; the shared furniture around it dominates what is read out The evidence, taken from the page as served How much of this page is its own content
Counted from the page as served, before any script runs. The rest is the menus, header and foot-of-page text that appear on other pages of the site as well.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: most of the page's text sits in its main content region rather than in the surrounding furniture. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fmedia%2Fsocial-media" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none" xmlns="http://w… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 1 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 1 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Social media H2 Social media H4 X (Twitter) level jumped H4 LinkedIn H4 Facebook H4 Instagram H4 Glassdoor H4 YouTube H4 Follow us H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT34 | fail | the page's own words are thin: 580 of its 601 words are shared navigation and footer, so it reads as 96% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 580 of its 601 words are shared navigation and footer, so it reads as 96% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/investors-and-reporting/results-centre 32.6% Very likely 8 out of 45
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 20 | |||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 20 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="How we're performing"/> | |||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Results centre H2 Results centre H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Finvestors-and-reporting%2Fresults-centre" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="n… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Results centre H2 Results centre H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-b85aabaf82-cta-d3dad6a9e2" data-cmp-data-layer="{"batcom-teaser-b85aabaf82-cta-d3dad6a9e2":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| CT34 | fail | the page's own words are thin: 589 of its 676 words are shared navigation and footer, so it reads as 89% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 589 of its 676 words are shared navigation and footer, so it reads as 89% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
▸https://www.bat.com/who-we-are/our-values 32.2% Very likely 8 out of 45
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Our values H2 Our values H3 We are BAT. Our values H4 Truly inclusive H4 Do the right thing H4 Love our consumer H4 Passion to win H4 Empowered through trust H4 Stronger together H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fwho-we-are%2Four-values" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none" xmlns="http:… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Our values H2 Our values H3 We are BAT. Our values H4 Truly inclusive H4 Do the right thing H4 Love our consumer H4 Passion to win H4 Empowered through trust H4 Stronger together H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-49b089c7c0-cta-7f6b2b5740" data-cmp-data-layer="{"batcom-teaser-49b089c7c0-cta-7f6b2b5740":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT34 | fail | the page's own words are thin: 588 of its 712 words are shared navigation and footer, so it reads as 83% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 588 of its 712 words are shared navigation and footer, so it reads as 83% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/who-we-are/our-people 54.1% Likely 8 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 199 | |||||||||||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 199 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="The quality of our people is a major reason why our Group continues to perform well – and why we are so excited about the future as we work together to deliver our purpose to build A Better Tomorrow."/> | |||||||||||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Our people H2 Our people H3 Our approach H3 (no text) H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fwho-we-are%2Four-people" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none" xmlns="http:… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Our people H2 Our people H3 Our approach H3 (no text) H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-77b0456b67-cta-7f6b2b5740" data-cmp-data-layer="{"batcom-teaser-77b0456b67-cta-7f6b2b5740":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/investors-and-reporting 32.2% Very likely 8 out of 45
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 1 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 1 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Investors & Reporting H2 Our latest public disclosures and investor information H2 2026 Half-Year Results H3 Current share price & latest results H3 Latest
2026 Half-Year Results H3 Key documents H3 Reports
2025 Combined Annual and Sustainability Report H3 Results
2025 Full-Year Results H2 Video: 2026 Half-Year Results Highlights H2 Investing in BAT H2 Investor factsheet H2 Sustainability H2 Important dates for investors H2 Investor presentations H2 News and filings H2 Reporting H2 Dividend information H2 Shareholder information H3 You might also be interested in H2 Video: 2026 Half-Year Results Highlights H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-633250c70d" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-7c62ff11c0" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7162a2146a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “read more”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “read more”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-1dcc9272a2-cta-e4d7ce1838" aria-label="Read more " href="/investors-and-reporting/investing-in-bat">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-acfb99ac12-cta-63179406b5" aria-label="Read more" href="/investors-and-reporting/investor-relations-fact-sheet">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-6d03522143-cta-df959e1d98" aria-label="Read more " href="/investors-and-reporting/sustainability-for-investors">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-29f81cce83-cta-d740c6a28b" aria-label="Read more " href="/investors-and-reporting/investor-presentations">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-b1d89ab852-cta-c14166f1ae" aria-label="Read more" href="/investors-and-reporting/news-and-filings">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-f4e3d4b84c-cta-a083eccb0a" aria-label="Read more " href="/investors-and-reporting/reporting">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-d7f81b0342-cta-5f1fd3ceea" aria-label="Read more " href="/investors-and-reporting/shareholder-information/dividend-information">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-374bb5bec4-cta-115c13f2bd" aria-label="Read more" href="/investors-and-reporting/shareholder-information">Read more</a> Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 1 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 1 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Investors & Reporting H2 Our latest public disclosures and investor information H2 2026 Half-Year Results H3 Current share price & latest results H3 Latest
2026 Half-Year Results H3 Key documents H3 Reports
2025 Combined Annual and Sustainability Report H3 Results
2025 Full-Year Results H2 Video: 2026 Half-Year Results Highlights H2 Investing in BAT H2 Investor factsheet H2 Sustainability H2 Important dates for investors H2 Investor presentations H2 News and filings H2 Reporting H2 Dividend information H2 Shareholder information H3 You might also be interested in H2 Video: 2026 Half-Year Results Highlights H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-1dcc9272a2-cta-e4d7ce1838" aria-label="Read more " href="/investors-and-reporting/investing-in-bat">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-acfb99ac12-cta-63179406b5" aria-label="Read more" href="/investors-and-reporting/investor-relations-fact-sheet">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-6d03522143-cta-df959e1d98" aria-label="Read more " href="/investors-and-reporting/sustainability-for-investors">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-29f81cce83-cta-d740c6a28b" aria-label="Read more " href="/investors-and-reporting/investor-presentations">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-b1d89ab852-cta-c14166f1ae" aria-label="Read more" href="/investors-and-reporting/news-and-filings">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-f4e3d4b84c-cta-a083eccb0a" aria-label="Read more " href="/investors-and-reporting/reporting">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-d7f81b0342-cta-5f1fd3ceea" aria-label="Read more " href="/investors-and-reporting/shareholder-information/dividend-information">Read more </a> <a class="cmp-teaser__action-link " id="batcom-teaser-374bb5bec4-cta-115c13f2bd" aria-label="Read more" href="/investors-and-reporting/shareholder-information">Read more</a> | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT34 | fail | the page's own words are thin: 587 of its 655 words are shared navigation and footer, so it reads as 87% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 587 of its 655 words are shared navigation and footer, so it reads as 87% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/who-we-are 32.0% Very likely 9 out of 45
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 200 | |||||||||||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 200 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="We are a leading multi-category consumer goods business. Our purpose is to build A Better Tomorrow™ by reducing the health impact of our business through offering a greater choice of enjoyable and les"/> | |||||||||||||||||||||
| CT09 | fail | 1 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 1 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Who we are H2 We are BAT H2 Our business H2 At a glance H2 Our values H2 Inspiring, challenging and diverse H2 Our leadership H2 Our corporate governance H2 Our history H2 We are proud of our partners H3 You might also be interested in H2 ‘The transformation is real, and we are delivering tangible change’ H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-633250c70d" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-7c62ff11c0" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7162a2146a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “read more”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “read more”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-24cd22baae-cta-ea0d09d838" aria-label="Read more" href="/who-we-are/at-a-glance">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-d36768fc88-cta-0f903e85db" aria-label="Read more" href="/who-we-are/our-values">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-b607356f34-cta-f7d592ed8b" aria-label="Read more" href="/who-we-are/our-leadership">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-ac340ef67c-cta-27a1d5960c" aria-label="Read more" href="/who-we-are/our-corporate-governance">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-5eda9d66e5-cta-de3811d9c4" aria-label="Read more" href="/who-we-are/our-history">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-4ef059dd04-cta-9f4c8378b8" data-cmp-data-layer="{"batcom-teaser-4ef059dd04-cta-9f4c8378b8":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/wh…Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 1 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 1 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Who we are H2 We are BAT H2 Our business H2 At a glance H2 Our values H2 Inspiring, challenging and diverse H2 Our leadership H2 Our corporate governance H2 Our history H2 We are proud of our partners H3 You might also be interested in H2 ‘The transformation is real, and we are delivering tangible change’ H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-24cd22baae-cta-ea0d09d838" aria-label="Read more" href="/who-we-are/at-a-glance">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-d36768fc88-cta-0f903e85db" aria-label="Read more" href="/who-we-are/our-values">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-b607356f34-cta-f7d592ed8b" aria-label="Read more" href="/who-we-are/our-leadership">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-ac340ef67c-cta-27a1d5960c" aria-label="Read more" href="/who-we-are/our-corporate-governance">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-5eda9d66e5-cta-de3811d9c4" aria-label="Read more" href="/who-we-are/our-history">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-4ef059dd04-cta-9f4c8378b8" data-cmp-data-layer="{"batcom-teaser-4ef059dd04-cta-9f4c8378b8":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/wh… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT34 | fail | the page's own words are thin: 587 of its 693 words are shared navigation and footer, so it reads as 82% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 587 of its 693 words are shared navigation and footer, so it reads as 82% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/who-we-are/at-a-glance 54.4% Likely 7 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 At a glance H2 At a glance H3 Our global business | BAT at-a-glance H4 As at 31 December 2025 (unless otherwise stated) H4 5 H4 £3,621m H4 34.1m H4 47,000+ H4 £25,610m H4 £11,279m H4 AAA H4 36 H3 In more detail H3 About BAT H3 Our purpose: to build A Better Tomorrow™ H3 Our brands H3 2025 results highlights H3 Our new 2030 sustainability targets H4 CLIMATE H4 NATURE H4 CIRCULARITY H4 COMMUNITIES H3 Forward-looking statements H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fwho-we-are%2Fat-a-glance" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none" xmlns="http… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 At a glance H2 At a glance H3 Our global business | BAT at-a-glance H4 As at 31 December 2025 (unless otherwise stated) H4 5 H4 £3,621m H4 34.1m H4 47,000+ H4 £25,610m H4 £11,279m H4 AAA H4 36 H3 In more detail H3 About BAT H3 Our purpose: to build A Better Tomorrow™ H3 Our brands H3 2025 results highlights H3 Our new 2030 sustainability targets H4 CLIMATE H4 NATURE H4 CIRCULARITY H4 COMMUNITIES H3 Forward-looking statements H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-9dd76c3564-cta-7f6b2b5740" data-cmp-data-layer="{"batcom-teaser-9dd76c3564-cta-7f6b2b5740":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/who-we-are/our-leadership/management-board 32.0% Very likely 9 out of 45
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 258 | |||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 258 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="BAT’s Management Board, chaired by our Chief Executive, is responsible for overseeing the implementation of the Group’s strategy and policies set by the Board, and for creating the framework for the day-to-day operation of the Group’s operating subsidiaries."/> | |||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Management Board H2 Management Board H2 Tadeu Marroco H2 Javed Iqbal H2 James Barrett H2 Luciano Comin H2 Zafar Khan H2 Dr Cora Koppe-Stahrenberg H2 Paul McCrory H2 Pascale Meulemeester H2 Fred Monteiro H2 Dr James Murphy H2 Johan Vandermeulen H2 David Waterfield H2 Kingsley Wheaton H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 2 link wording(s) are used for more than one destination, so the same words point at different pages: “full biography”, “linkedin”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 2 link wording(s) are used for more than one destination, so the same words point at different pages: “full biography”, “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-a4f6e679e5-cta-92d3e370bc" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/tadeu-marroco">Full biography</a> <a class="cmp-teaser__action-link " id="batcom-teaser-523fefffce-cta-8c5fde308e" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/javed-iqbal">Full biography</a> <a class="cmp-teaser__action-link " id="batcom-teaser-bd282d5a08-cta-737c94c60b" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/james-barrett">Full biography</a> <a class="cmp-teaser__action-link " id="batcom-teaser-698dec5788-cta-72fdd31ce3" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/luciano-comin">Full biography</a> <a class="cmp-teaser__action-link " id="batcom-teaser-176df05371-cta-14fd3800e7" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/zafar-khan">Full biography</a> <a class="cmp-teaser__action-link " id="batcom-teaser-dc1cea712a-cta-ad0b156468" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/dr-cora-koppe-stahrenberg">Full biography</a> <a class="cmp-teaser__action-link " id="batcom-teaser-a8b332acc4-cta-fa5627b7f3" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/paul-mccrory">Full biography</a> <a class="cmp-teaser__action-link " id="batcom-teaser-32b4372a34-cta-d6b7415736" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/pascale-meulemeester">Full biography</a> <a class="cmp-teaser__action-link " id="batcom-teaser-314a67217c-cta-3bc0533531" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/fred-monteiro">Full biography</a> <a class="cmp-teaser__action-link " id="batcom-teaser-4e1ca7c49f-cta-b4b5ea4b36" aria-label="Full biography" href="/who-we-are/our-leadership/management-board/dr-james-murphy">Full biography</a> Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Management Board H2 Management Board H2 Tadeu Marroco H2 Javed Iqbal H2 James Barrett H2 Luciano Comin H2 Zafar Khan H2 Dr Cora Koppe-Stahrenberg H2 Paul McCrory H2 Pascale Meulemeester H2 Fred Monteiro H2 Dr James Murphy H2 Johan Vandermeulen H2 David Waterfield H2 Kingsley Wheaton H3 You might also be interested in H2 2025 People and Culture Report H4 Who we are level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-078747bd7a-cta-7f6b2b5740" data-cmp-data-layer="{"batcom-teaser-078747bd7a-cta-7f6b2b5740":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT34 | fail | the page's own words are thin: 589 of its 718 words are shared navigation and footer, so it reads as 82% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 589 of its 718 words are shared navigation and footer, so it reads as 82% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
▸https://www.bat.com/investors-and-reporting/investing-in-bat 54.4% Likely 7 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 3 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 3 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Investing in BAT H3 Investing in BAT level jumped H4 Quick links H2 Transformation Driving Sustainable Growth H2 (no text) H2 Dynamic Business Making Active Choices for the Future H2 Continuing Our Track Record of Delivery H2 Data centre H2 Transformation Driving Sustainable Growth H2 Creating Sustainable Value for our Stakeholders H2 Dynamic Business Making Active Choices for the Future H2 Continuing our track record of delivery H2 Data centre H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Finvestors-and-reporting%2Finvesting-in-bat" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill=… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 3 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 3 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Investing in BAT H3 Investing in BAT level jumped H4 Quick links H2 Transformation Driving Sustainable Growth H2 (no text) H2 Dynamic Business Making Active Choices for the Future H2 Continuing Our Track Record of Delivery H2 Data centre H2 Transformation Driving Sustainable Growth H2 Creating Sustainable Value for our Stakeholders H2 Dynamic Business Making Active Choices for the Future H2 Continuing our track record of delivery H2 Data centre H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-dc82d2ffb4-cta-a3eb493b26" aria-label="Read more" href="#transformation">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-892aa2e7b2-cta-1bab31493c" aria-label="Read more" href="#reducing">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-e194e9f0d9-cta-2cc8dd77e8" aria-label="Read more" href="#creating">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-4c93efe6c8-cta-2c39208e1d" aria-label="Read more" href="#continuing">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-143ba340e5-cta-0621cea7e2" aria-label="Read more" href="#data">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-d47559cd75-cta-d3dad6a9e2" data-cmp-data-layer="{"batcom-teaser-d47559cd75-cta-d3dad6a9e2":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/who-we-are/our-business 53.6% Likely 8 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 1 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 1 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H3 We are BAT H4 We are a leading global company H2 We are a multi-category business H2 We are shaping the future H2 We are responsible H2 We are partners H2 We are proud of our people H3 You might also be interested in H2 ‘The transformation is real, and we are delivering tangible change’ H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT10 | fail | the page carries no main heading, so nothing on it states what the page is about | |||||||||||||||||||
What we check The page carries a main heading (the H1), its own statement of what the page is about. What good looks like The page carries a single main heading stating what it is about. Why it matters With no main heading the page never states its own subject, so the sections beneath it hang from nothing. What we found on this page the page carries no main heading, so nothing on it states what the page is about The evidence, taken from the page as served The headings on this page, as served
The page was read for a top-level heading. These are the headings it does carry. <h3 class="batcom-hero-our-business__top-title batcom-hero-our-business__text">We are BAT</h3> <h4 class="batcom-hero-our-business__bottom-title batcom-hero-our-business__text">We are a leading global company</h4> <h2 class="cmp-teaser__title"> We are a multi-category business </h2> <h2 class="cmp-teaser__title"> We are shaping the future </h2> <h2 class="cmp-teaser__title"> We are responsible </h2> <h2 class="cmp-teaser__title"> We are partners </h2> <h2 class="cmp-teaser__title"> We are proud of our people </h2> <h3><span style="font-weight: normal;">You might also be interested in</span></h3> <h2 class="cmp-teaser__title"> ‘The transformation is real, and we are delivering tangible change’ </h2> <h2 class="batcom-linklist__title">Related Articles</h2> …and 11 more like it. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page carries a single main heading stating what it is about. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT04 | warning | the page shows the year 2025 in its text, but nothing on the page states it as the date of the content | |||||||||||||||||||
What we check The page carries a date, ideally written into the markup rather than sitting only in the text. What good looks like The page carries a date, ideally written into the markup rather than sitting only in the text. Why it matters An undated page gives nothing to weigh its age against, so old and new material are treated alike. What we found on this page the page shows the year 2025 in its text, but nothing on the page states it as the date of the content The evidence, taken from the page as served The head of this page, as served
The head of the page was read for a tag stating the date of the content. The year shown above was found in the page's own text. <head> <script defer="defer" type="text/javascript" src="https://rum.hlx.page/.rum/@adobe/helix-rum-js@%5E2/dist/rum-standalone.js" data-routing="env=prod,tier=publish,ams=British American Tobacco"/> <script src="https://cdn.cookielaw.org/scripttemplates/otSDKStub.js" async="" data-domain-script="5fc350fc-73ff-46f2-9463-37d2cf7f5f1a" charset="UTF-8"/> <script> (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push ({'gtm.start': new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0], j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src= 'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f); })(window,document,'script','dataLayer','GTM-TMH7W3WX');</script> <meta charset="UTF-8"/> <title> British American Tobacco - Our business </title> <meta name="keywords" content="transforming tobacco,We Are BAT,heritage,company brochure,our people,history,reduced risk products,supporting farmers"/> <meta name="description" content="We are BAT, a leading, multi-category consumer goods business. Our purpose is to create A Better Tomorrow™ through building a smokeless world."/> <meta name="template" content="contentpage-template"/> <meta name="… | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-ced40bf7b7" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-7d53ca0f29" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-0e96d25f94" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS10 | fail | the page's first heading is not its top-level heading, so the structure starts part way down; the heading structure skips a level in 1 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the page's first heading is not its top-level heading, so the structure starts part way down; the heading structure skips a level in 1 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H3 We are BAT level jumped H4 We are a leading global company H2 We are a multi-category business H2 We are shaping the future H2 We are responsible H2 We are partners H2 We are proud of our people H3 You might also be interested in H2 ‘The transformation is real, and we are delivering tangible change’ H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-fe2ad99f89-cta-9f4c8378b8" data-cmp-data-layer="{"batcom-teaser-fe2ad99f89-cta-9f4c8378b8":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/wh… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/brands-and-innovation/product-categories 32.2% Very likely 8 out of 45
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Product categories H2 Product categories H3 Modern Oral H3 Vapour H3 Heated Product H3 Traditional Oral H3 Combustibles H3 You might also be interested in H2 A decade of portfolio transformation at BAT H4 Brands & Innovation level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fbrands-and-innovation%2Fproduct-categories" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill=… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Product categories H2 Product categories H3 Modern Oral H3 Vapour H3 Heated Product H3 Traditional Oral H3 Combustibles H3 You might also be interested in H2 A decade of portfolio transformation at BAT H4 Brands & Innovation level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-99f90a1d60-cta-aba2cae2de" data-cmp-data-layer="{"batcom-teaser-99f90a1d60-cta-aba2cae2de":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/me… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT34 | fail | the page's own words are thin: 589 of its 709 words are shared navigation and footer, so it reads as 83% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 589 of its 709 words are shared navigation and footer, so it reads as 83% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/sustainability-and-esg/climate 53.6% Likely 8 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MX01 | warning | the site took 810ms to start answering; beyond about 345ms an automated reader is liable to give up first | |||||||||||||||||||
What we check How quickly the page starts to reply, measured against a 345 millisecond first-byte threshold. What good looks like The page starts to reply promptly, within the first-byte threshold. Why it matters Past the waiting limit the answer is assembled from stale or third-party material instead of from your page. What we found on this page the site took 810ms to start answering; beyond about 345ms an automated reader is liable to give up first The evidence, taken from the page as served How long this page took to start answering
Timed from the request leaving our checker to the first byte coming back, on the page as served. The 345ms figure is the point beyond which an automated reader is liable to give up.
| |||||||||||||||||||||
| CT09 | fail | 3 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 3 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Climate H2 Climate H4 2030 targets: level jumped H4 How we’ll get there - working towards Net Zero across our value chain by 2050 H4 A roadmap to Net Zero H4 Our CDP Publications H2 Combined Annual and Sustainability Report H2 Sustainability Policies, Procedure and Standards H2 Sustainability Performance Data Book H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 2 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”, “read more”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 2 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”, “read more”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-192791002c-cta-8af66a1497" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-ebda10df8b-cta-b556c7db2a" aria-label="Read more" href="/sustainability-and-esg/sustainability-governance/sustainability-policies-procedures-and-standards">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-47a02f3fa2-cta-f4e554ece9" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting">Read more</a> <a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fsustainability-and-esg%2Fclimate" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none" xml… <a class="cmp-teaser__action-link " id="batcom-teaser-62f7de17ac-cta-0b630cf0e5" data-cmp-data-layer="{"batcom-teaser-62f7de17ac-cta-0b630cf0e5":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in…<a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 3 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 3 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Climate H2 Climate H4 2030 targets: level jumped H4 How we’ll get there - working towards Net Zero across our value chain by 2050 H4 A roadmap to Net Zero H4 Our CDP Publications H2 Combined Annual and Sustainability Report H2 Sustainability Policies, Procedure and Standards H2 Sustainability Performance Data Book H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-192791002c-cta-8af66a1497" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-ebda10df8b-cta-b556c7db2a" aria-label="Read more" href="/sustainability-and-esg/sustainability-governance/sustainability-policies-procedures-and-standards">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-47a02f3fa2-cta-f4e554ece9" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-62f7de17ac-cta-0b630cf0e5" data-cmp-data-layer="{"batcom-teaser-62f7de17ac-cta-0b630cf0e5":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/sustainability-and-esg/communities 54.4% Likely 7 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Communities H2 Communities H3 How we’ll get there H4 Employees H4 Farmers H4 Non-tobacco suppliers H3 Communities identified in our Double Materiality Assessment2 H2 Combined Annual and Sustainability Report H2 Sustainability Performance Data Book H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 2 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”, “read more”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 2 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”, “read more”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-ce9b9a0265-cta-cf56fb4074" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-137f5d1140-cta-72fbc189dd" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting">Read more</a> <a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fsustainability-and-esg%2Fcommunities" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none"… <a class="cmp-teaser__action-link " id="batcom-teaser-6b5ed9fcad-cta-0b630cf0e5" data-cmp-data-layer="{"batcom-teaser-6b5ed9fcad-cta-0b630cf0e5":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in…<a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Communities H2 Communities H3 How we’ll get there H4 Employees H4 Farmers H4 Non-tobacco suppliers H3 Communities identified in our Double Materiality Assessment2 H2 Combined Annual and Sustainability Report H2 Sustainability Performance Data Book H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-ce9b9a0265-cta-cf56fb4074" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-137f5d1140-cta-72fbc189dd" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-6b5ed9fcad-cta-0b630cf0e5" data-cmp-data-layer="{"batcom-teaser-6b5ed9fcad-cta-0b630cf0e5":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/investors-and-reporting/reporting 31.0% Very likely 10 out of 45
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 200 | |||||||||||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 200 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="In this section you can access key information related to our financial, ESG and other statutory reporting. You can view and download all our latest reports which detail how and why we are building A "/> | |||||||||||||||||||||
| CT09 | fail | 1 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 1 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H2 (no text) H3 Reporting H2 Annual reporting - how we're performing H2 Focus Reporting H2 Sustainability Reporting H2 Gender Pay Report H2 Conflict Minerals Report H2 Modern Slavery Statement H3 You might also be interested in H2 Video: 2026 Half-Year Results Highlights H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT10 | fail | the page carries no main heading, so nothing on it states what the page is about | |||||||||||||||||||
What we check The page carries a main heading (the H1), its own statement of what the page is about. What good looks like The page carries a single main heading stating what it is about. Why it matters With no main heading the page never states its own subject, so the sections beneath it hang from nothing. What we found on this page the page carries no main heading, so nothing on it states what the page is about The evidence, taken from the page as served The headings on this page, as served
The page was read for a top-level heading. These are the headings it does carry. <h2 class="cmp-teaser__title"> </h2> <h3><b>Reporting</b></h3> <h2 class="cmp-teaser__title"> Annual reporting - how we're performing </h2> <h2 class="cmp-teaser__title"> Focus Reporting </h2> <h2 class="cmp-teaser__title"> Sustainability Reporting </h2> <h2 class="cmp-teaser__title"> Gender Pay Report </h2> <h2 class="cmp-teaser__title"> Conflict Minerals Report </h2> <h2 class="cmp-teaser__title"> Modern Slavery Statement </h2> <h3><span style="font-weight: normal;">You might also be interested in</span></h3> <h2 class="cmp-teaser__title"> Video: 2026 Half-Year Results Highlights </h2> …and 12 more like it. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page carries a single main heading stating what it is about. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-ced40bf7b7" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-7d53ca0f29" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-0e96d25f94" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “read more”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “read more”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-e8354b7299-cta-59ed93144b" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-f66c260695-cta-52dd24a8da" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting/focus-reporting">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-6a385c79c0-cta-ff760f083f" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-d99bc3bf62-cta-1d687aa1e0" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting/people-and-culture-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-296c9667c0-cta-316afbfa20" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting/conflict-minerals-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-bdeb85025e-cta-a1a1e4e60a" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting/modern-slavery-statement">Read more</a> Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the page's first heading is not its top-level heading, so the structure starts part way down; the heading structure skips a level in 1 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the page's first heading is not its top-level heading, so the structure starts part way down; the heading structure skips a level in 1 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H2 (no text) level jumped H3 Reporting H2 Annual reporting - how we're performing H2 Focus Reporting H2 Sustainability Reporting H2 Gender Pay Report H2 Conflict Minerals Report H2 Modern Slavery Statement H3 You might also be interested in H2 Video: 2026 Half-Year Results Highlights H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-e8354b7299-cta-59ed93144b" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-f66c260695-cta-52dd24a8da" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting/focus-reporting">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-6a385c79c0-cta-ff760f083f" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-d99bc3bf62-cta-1d687aa1e0" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting/people-and-culture-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-296c9667c0-cta-316afbfa20" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting/conflict-minerals-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-bdeb85025e-cta-a1a1e4e60a" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting/modern-slavery-statement">Read more</a> | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT34 | fail | the page's own words are thin: 587 of its 639 words are shared navigation and footer, so it reads as 88% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 587 of its 639 words are shared navigation and footer, so it reads as 88% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/brands-and-innovation/building-brands 54.1% Likely 8 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 200 | |||||||||||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 200 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="BAT is a leading consumer-centric, multi-category consumer goods business dedicated to stimulating the senses of adult consumers worldwide. Our portfolio reflects our commitment to meeting the evolvin"/> | |||||||||||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Building brands H2 Building brands H3 Our Strategic Portfolio H3 (no text) H3 Other international and local cigarette brands H3 (no text) H3 You might also be interested in H2 A decade of portfolio transformation at BAT H4 Brands & Innovation level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fbrands-and-innovation%2Fbuilding-brands" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="no… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Building brands H2 Building brands H3 Our Strategic Portfolio H3 (no text) H3 Other international and local cigarette brands H3 (no text) H3 You might also be interested in H2 A decade of portfolio transformation at BAT H4 Brands & Innovation level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-f1de2c56bd-cta-aba2cae2de" data-cmp-data-layer="{"batcom-teaser-f1de2c56bd-cta-aba2cae2de":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/me… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/brands-and-innovation/driving-innovation 54.4% Likely 7 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Driving innovation H2 Driving innovation H3 You might also be interested in H2 A decade of portfolio transformation at BAT H4 Brands & Innovation level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fbrands-and-innovation%2Fdriving-innovation" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill=… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Driving innovation H2 Driving innovation H3 You might also be interested in H2 A decade of portfolio transformation at BAT H4 Brands & Innovation level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-8ef11326dc-cta-aba2cae2de" data-cmp-data-layer="{"batcom-teaser-8ef11326dc-cta-aba2cae2de":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/me… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/sustainability-and-esg/nature 54.4% Likely 7 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 3 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 3 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Nature H2 Nature H4 Our ambition - contributing to a Nature Positive1 future. level jumped H4 2030 targets H4 How we’ll get there H5 1. AVOID H5 2. REDUCE H5 3. RESTORE AND REGENERATE H5 4. TRANSFORM H4 Our CDP Publications: H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fsustainability-and-esg%2Fnature" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none" xmln… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 3 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 3 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Nature H2 Nature H4 Our ambition - contributing to a Nature Positive1 future. level jumped H4 2030 targets H4 How we’ll get there H5 1. AVOID H5 2. REDUCE H5 3. RESTORE AND REGENERATE H5 4. TRANSFORM H4 Our CDP Publications: H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-44cf4c12a3-cta-0b630cf0e5" data-cmp-data-layer="{"batcom-teaser-44cf4c12a3-cta-0b630cf0e5":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/sustainability-and-esg/circularity 54.4% Likely 7 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 4 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 4 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Circularity H2 Circularity H4 2030 targets level jumped H4 How we’ll get there H4 Our circularity strategy H2 Combined Annual and Sustainability Report H2 Sustainability Policies, Procedure and Standards H2 Sustainability Performance Data Book H5 Notes: level jumped H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 2 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”, “read more”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 2 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”, “read more”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-856c89909d-cta-d5ba209413" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-2f8c9db254-cta-3b104c3a06" aria-label="Read more" href="/sustainability-and-esg/sustainability-governance/sustainability-policies-procedures-and-standards">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-b0dca5d902-cta-3f7f07fd34" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting">Read more</a> <a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fsustainability-and-esg%2Fcircularity" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none"… <a class="cmp-teaser__action-link " id="batcom-teaser-72d6c0e1c7-cta-0b630cf0e5" data-cmp-data-layer="{"batcom-teaser-72d6c0e1c7-cta-0b630cf0e5":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in…<a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 4 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 4 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Circularity H2 Circularity H4 2030 targets level jumped H4 How we’ll get there H4 Our circularity strategy H2 Combined Annual and Sustainability Report H2 Sustainability Policies, Procedure and Standards H2 Sustainability Performance Data Book H5 Notes: level jumped H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Sustainability & ESG level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-856c89909d-cta-d5ba209413" aria-label="Read more" href="/investors-and-reporting/reporting/combined-annual-and-esg-report">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-2f8c9db254-cta-3b104c3a06" aria-label="Read more" href="/sustainability-and-esg/sustainability-governance/sustainability-policies-procedures-and-standards">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-b0dca5d902-cta-3f7f07fd34" aria-label="Read more" href="/investors-and-reporting/reporting/sustainability-reporting">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-72d6c0e1c7-cta-0b630cf0e5" data-cmp-data-layer="{"batcom-teaser-72d6c0e1c7-cta-0b630cf0e5":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/investors-and-reporting/reporting/sustainability-reporting 32.2% Very likely 9 out of 45
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 15 | |||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 15 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="Why it matters."/> | |||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Sustainability Reporting H2 Sustainability Reporting H3 Sustainability Reporting 2025 H4 Sustainability Performance Data Book 2025 H3 Previous sustainability reporting H3 (no text) H3 Group stakeholder dialogue reports H3 (no text) H3 Company stakeholder dialogue reports H3 (no text) H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Finvestors-and-reporting%2Freporting%2Fsustainability-reporting" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewb… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Sustainability Reporting H2 Sustainability Reporting H3 Sustainability Reporting 2025 H4 Sustainability Performance Data Book 2025 H3 Previous sustainability reporting H3 (no text) H3 Group stakeholder dialogue reports H3 (no text) H3 Company stakeholder dialogue reports H3 (no text) H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Investors & Reporting level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-c52bb6d667-cta-d3dad6a9e2" data-cmp-data-layer="{"batcom-teaser-c52bb6d667-cta-d3dad6a9e2":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| CT34 | fail | the page's own words are thin: 589 of its 712 words are shared navigation and footer, so it reads as 85% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 589 of its 712 words are shared navigation and footer, so it reads as 85% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
▸https://www.bat.com/strategy-and-purpose/reducing-harm 52.7% Likely 9 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 195 | |||||||||||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 195 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="Tobacco harm reduction is a public health strategy that is about minimising the negative health impact of smoking. It is core to our delivery of Building a Smokeless World and A Better Tomorrow™."/> | |||||||||||||||||||||
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Reducing harm H2 Reducing harm H3 Our Approach at a Glance H3 Applying Best Practice Research Standards H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Strategy & Purpose level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fstrategy-and-purpose%2Freducing-harm" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none"… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Reducing harm H2 Reducing harm H3 Our Approach at a Glance H3 Applying Best Practice Research Standards H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Strategy & Purpose level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-6c067c1b39-cta-bf7b6b0f00" data-cmp-data-layer="{"batcom-teaser-6c067c1b39-cta-bf7b6b0f00":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| MS14 | fail | 1 video(s) on the page with no captions on the video and no written transcript on the page: nothing spoken is available as text in any form | |||||||||||||||||||
What we check Whether each video carries captions of its own, timed text on the video itself, judged together with whether the page holds a written transcript. What good looks like Captions carried on the video. A WARNING means the video has no captions but a written transcript on the page holds the words. A FAIL means neither exists, so nothing spoken is available as text in any form; that total absence is marked here once, not spread across two findings. Why it matters Spoken content that exists only as sound is invisible to a reader, human or machine: nothing said in the video can be found, read or quoted. What we found on this page 1 video(s) on the page with no captions on the video and no written transcript on the page: nothing spoken is available as text in any form The evidence, taken from the page as served Each video on the page, as served: its code, source file and caption track
No caption track on the video and no transcript link on the page. <video class="batcom-video__video" src="/content/dam/batcom/global/video-files/main-site/reducing-harm/Advancing_THR_1-new.m3u8" poster="/content/dam/batcom/global/video-files/main-site/reducing-harm/Advancing_THR_1.jpg" preload="none" playsinline="" controlslist="nofullscreen"> </video> Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: captions carried on the video. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/brands-and-innovation 32.2% Very likely 8 out of 45
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 1 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 1 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Brands & Innovation H2 Building purposeful brands and innovative products H2 Building brands H2 Velo H2 Vuse H2 glo H2 Driving innovation H2 Manufacturing H2 Working with retailers H2 Regulation H2 Product categories H3 You might also be interested in H2 A decade of portfolio transformation at BAT H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-633250c70d" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-7c62ff11c0" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7162a2146a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 2 link wording(s) are used for more than one destination, so the same words point at different pages: “explore more”, “read more”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 2 link wording(s) are used for more than one destination, so the same words point at different pages: “explore more”, “read more”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-acb0680cd9-cta-6454865b94" aria-label="Read more" href="/brands-and-innovation/velo">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-7eb627e16c-cta-0456efb7a2" aria-label="Read more" href="/brands-and-innovation/vuse">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-b1be281ce6-cta-1b26b02c4b" aria-label="Read more" href="/brands-and-innovation/glo">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-84fbf263ab-cta-0f43ac5741" aria-label="Explore more" href="/brands-and-innovation/driving-innovation">Explore more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-ea8ebac8d3-cta-f353ce8f5b" aria-label="Read more" href="/brands-and-innovation/manufacturing">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-955ee3d2f4-cta-56ca450cee" aria-label="Read more" href="/brands-and-innovation/working-with-retailers">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-862464e140-cta-e18658123c" aria-label="Read more" href="/strategy-and-purpose/progressive-regulation">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-6b7fa2eb9e-cta-3816e205b7" aria-label="Explore more" href="/brands-and-innovation/product-categories">Explore more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-198d7c466f-cta-ecffaa3ad7" data-cmp-data-layer="{"batcom-teaser-198d7c466f-cta-ecffaa3ad7":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/me…Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 1 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 1 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Brands & Innovation H2 Building purposeful brands and innovative products H2 Building brands H2 Velo H2 Vuse H2 glo H2 Driving innovation H2 Manufacturing H2 Working with retailers H2 Regulation H2 Product categories H3 You might also be interested in H2 A decade of portfolio transformation at BAT H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-acb0680cd9-cta-6454865b94" aria-label="Read more" href="/brands-and-innovation/velo">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-7eb627e16c-cta-0456efb7a2" aria-label="Read more" href="/brands-and-innovation/vuse">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-b1be281ce6-cta-1b26b02c4b" aria-label="Read more" href="/brands-and-innovation/glo">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-ea8ebac8d3-cta-f353ce8f5b" aria-label="Read more" href="/brands-and-innovation/manufacturing">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-955ee3d2f4-cta-56ca450cee" aria-label="Read more" href="/brands-and-innovation/working-with-retailers">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-862464e140-cta-e18658123c" aria-label="Read more" href="/strategy-and-purpose/progressive-regulation">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-198d7c466f-cta-ecffaa3ad7" data-cmp-data-layer="{"batcom-teaser-198d7c466f-cta-ecffaa3ad7":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/me… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT34 | fail | the page's own words are thin: 587 of its 673 words are shared navigation and footer, so it reads as 84% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 587 of its 673 words are shared navigation and footer, so it reads as 84% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/brands-and-innovation/manufacturing 54.4% Likely 7 out of 44
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Manufacturing H2 Manufacturing H3 Delivering on Decarbonisation H3 Inside a cigarette factory H3 Product Quality and Safety H3 You might also be interested in H2 A decade of portfolio transformation at BAT H4 Brands & Innovation level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “linkedin”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fbrands-and-innovation%2Fmanufacturing" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fill="none… <a href="https://www.linkedin.com/company/british-american-tobacco" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg role="img" viewbox="0 0 24 24" xmlns="http://www.w3.org/2000/svg" id="Follow-Us-LinkedIn" height="24" width="24"> <ti… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Manufacturing H2 Manufacturing H3 Delivering on Decarbonisation H3 Inside a cigarette factory H3 Product Quality and Safety H3 You might also be interested in H2 A decade of portfolio transformation at BAT H4 Brands & Innovation level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-be7a00ecde-cta-aba2cae2de" data-cmp-data-layer="{"batcom-teaser-be7a00ecde-cta-aba2cae2de":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/me… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
▸https://www.bat.com/science-and-research/reduced-risk-products 54.4% Likely 7 out of 44
| Check | Result | Evidence | |
|---|---|---|---|
| CT09 | fail | 2 skipped heading level(s) in document order | |
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 2 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Reduced-risk products H2 Reduced-risk products H3 Vapour Products H3 Heated Products H3 Modern Oral Products H3 Product pipeline H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Science & Research level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-872a9f0ff6" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-ff1789fc0a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7dbdc62b53" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||
| MS08 | fail | 4 link wording(s) are used for more than one destination, so the same words point at different pages: “heated products”, “linkedin”, “modern oral products” and 1 more; the links are listed in the view | |
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 4 link wording(s) are used for more than one destination, so the same words point at different pages: “heated products”, “linkedin”, “modern oral products” and 1 more; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a itemprop="url" class="cmp-navigationcorp__item-link" title="Modern oral products" href="/brands-and-innovation/product-categories/modern-oral-products"> <span class="cmp-navigationcorp__item-link-title">Modern oral products</span></a> <a itemprop="url" class="cmp-navigationcorp__item-link" title="Vapour products" href="/brands-and-innovation/product-categories/vapour-products"> <span class="cmp-navigationcorp__item-link-title">Vapour products</span></a> <a itemprop="url" class="cmp-navigationcorp__item-link" title="Heated Products" href="/brands-and-innovation/product-categories/heated-products"> <span class="cmp-navigationcorp__item-link-title">Heated Products</span></a> <a href="/science-and-research/modern-oral" title="Modern Oral" target="_self" data-link-type="internal">Modern Oral Products</a> <a href="/science-and-research/vapour" title="Vapour" target="_self" data-link-type="internal">Vapour Products</a> <a href="/science-and-research/heated-tobacco" title="Heated Products" target="_self" data-link-type="internal">Heated Products</a> <a href="/brands-and-innovation/product-categories/vapour-products" title="Vapour products" target="_self" data-link-type="internal">Vapour Products</a> <a href="/brands-and-innovation/product-categories/heated-products" title="Heated Products" target="_self" data-link-type="internal">Heated Products</a> <a href="/brands-and-innovation/product-categories/modern-oral-products" title="Modern and Traditional Oral Products" target="_self" data-link-type="internal">Modern Oral Products</a> <a href="https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.bat.com%2Fscience-and-research%2Freduced-risk-products" class="socialmedia__link" target="_blank" rel="noopener" aria-label="LinkedIn"> <span class="socialmedia__svg" aria-hidden="true"><svg width="22" height="23" viewbox="0 0 22 23" fil… Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||
| MS10 | fail | the heading structure skips a level in 2 place(s), so the order does not follow | |
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 2 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Reduced-risk products H2 Reduced-risk products H3 Vapour Products H3 Heated Products H3 Modern Oral Products H3 Product pipeline H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H4 Science & Research level jumped H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||
| MS09 | warning | link wording that does not say where it leads: “read more” | |
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-ada225dd57-cta-51c70db8c9" data-cmp-data-layer="{"batcom-teaser-ada225dd57-cta-51c70db8c9":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||
| MS13 | warning | 1 of 1 iframes untitled | |
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning. Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||
▸https://www.bat.com/sustainability-and-esg 32.0% Very likely 9 out of 45
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CT32 | warning | length 177 | |||||||||||||||||||
What we check The meta description's length is usable, between 50 and 170 characters. What good looks like The meta description's length is usable, long enough to say something and short enough not to be cut. Why it matters Too short says nothing and too long gets cut, so the summary field is spent without doing its job. What we found on this page length 177 The evidence, taken from the page as served The page's description, as served
<meta name="description" content="Sustainability is a core part of our Group transformation strategy. As we work towards our vision of Building a Smokeless World, we recognise that we must transform responsibly."/> | |||||||||||||||||||||
| CT09 | fail | 1 skipped heading level(s) in document order | |||||||||||||||||||
What we check Heading levels step down one at a time, rather than jumping from an H2 straight to an H5. What good looks like Heading levels step down one at a time, without jumping a level. Why it matters A skipped level breaks the trail that says which section a passage belongs to, so it is filed under the wrong one. What we found on this page 1 skipped heading level(s) in document order The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Sustainability & ESG H2 Integrating sustainability for A Better Tomorrow™ H2 Our sustainability strategy H2 Setting our sustainability strategic priorities H2 Advancing our climate transition H2 Adopting technology in support of nature H2 Embedding circularity across our products H2 Promoting social and economic development H2 A strong framework and culture of integrity H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: heading levels step down one at a time, without jumping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT35 | warning | the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template | |||||||||||||||||||
What we check The copyright line at the foot of the page, and the year it states. What good looks like The copyright line states the current year. Why it matters A copyright year in the past reads as a page nobody has attended to, and it is the usual sign of a template copied from somewhere else. Either way an AI treats the content as stale and can present old material as current. What we found on this page the page carries no copyright line in the content as served, so there is nothing on it to date it by, and an absent copyright line is one of the marks of a copied template The evidence, taken from the page as served The foot of this page, as served
The foot of the page was read for a copyright line. <footer id="batcom-experiencefragment-633250c70d" role="contentinfo" class="cmp-experiencefragment cmp-experiencefragment--footer"> <div class="xf-content-height"> <div id="container-7c62ff11c0" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class="batcom-container container responsivegrid batcom-content"> <div id="container-7162a2146a" class="cmp-container " data-nc="Container" data-nc-params-container="{ "visibilityRestriction": "DEFAULT", "authenticationCookie" : "", "newsletterSignedUpCookie" : "newsletterSignedUp", "hasFrontendLogic" : false, "scrollArrowPreviousLabel": "Scroll left", "scrollArrowNextLabel": "Scroll right" }"> <div class=… | |||||||||||||||||||||
| MS08 | fail | 1 link wording(s) are used for more than one destination, so the same words point at different pages: “read more”; the links are listed in the view | |||||||||||||||||||
What we check Links that share exactly the same wording on one page while pointing at different destinations. What good looks like Links that share the same wording lead to the same destination. Why it matters The wording no longer settles the target, so the wrong one gets followed at the point where the choice matters. What we found on this page 1 link wording(s) are used for more than one destination, so the same words point at different pages: “read more”; the links are listed in the view The evidence, taken from the page as served Links that share the same text but point to different destinations, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-299b24920d-cta-f4c846c716" aria-label="Read more" href="/sustainability-and-esg/materiality">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-32407d7ede-cta-c5d79e9e60" aria-label="Read more" href="/sustainability-and-esg/climate">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-4f6f66d7d2-cta-854dd11e55" aria-label="Read more" href="/sustainability-and-esg/nature">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-c81ac67084-cta-0165cfdf7a" aria-label="Read more" href="/sustainability-and-esg/circularity">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-a0a97faab9-cta-61edadb49a" aria-label="Read more" href="/sustainability-and-esg/communities">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-68c6e69072-cta-c0161c8c65" aria-label="Read more" href="/sustainability-and-esg/sustainability-governance">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-fe169a5262-cta-a72d52f36f" data-cmp-data-layer="{"batcom-teaser-fe169a5262-cta-a72d52f36f":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in…Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: links that share the same wording lead to the same destination. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS10 | fail | the heading structure skips a level in 1 place(s), so the order does not follow | |||||||||||||||||||
What we check A page's headings start at the top level and step down without skipping a level. What good looks like The page's headings start at the top level and step down without skipping a level. Why it matters Content under a skipped level has no clear parent section, so structure taken from the page misplaces its text. What we found on this page the heading structure skips a level in 1 place(s), so the order does not follow The evidence, taken from the page as served The page's heading outline, in document order (a jump in level is marked)
H1 Sustainability & ESG H2 Integrating sustainability for A Better Tomorrow™ H2 Our sustainability strategy H2 Setting our sustainability strategic priorities H2 Advancing our climate transition H2 Adopting technology in support of nature H2 Embedding circularity across our products H2 Promoting social and economic development H2 A strong framework and culture of integrity H3 You might also be interested in H2 Omni™ – Forward Thinking for a Smokeless World H2 Related Articles H4 Follow us level jumped H4 Current UK share price H4 Quick Links H4 Modern Slavery Statement H4 Gender Pay Report H4 Combined Annual and Sustainability Report H4 Standards of Business Conduct H4 Omni™ – Forward Thinking for a Smokeless World H4 Share price data H4 Site map H4 Contact us
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's headings start at the top level and step down without skipping a level. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| MS09 | warning | link wording that does not say where it leads: “read more” | |||||||||||||||||||
What we check Link wording that names no destination, phrases such as click here, read more and details. What good looks like Link wording names its destination, rather than saying click here or read more. Why it matters The destination is stated nowhere on the link itself, so the choice rests entirely on the text around it. What we found on this page link wording that does not say where it leads: “read more” The evidence, taken from the page as served The links whose text names no destination, as served
<a class="cmp-teaser__action-link " id="batcom-teaser-299b24920d-cta-f4c846c716" aria-label="Read more" href="/sustainability-and-esg/materiality">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-32407d7ede-cta-c5d79e9e60" aria-label="Read more" href="/sustainability-and-esg/climate">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-4f6f66d7d2-cta-854dd11e55" aria-label="Read more" href="/sustainability-and-esg/nature">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-c81ac67084-cta-0165cfdf7a" aria-label="Read more" href="/sustainability-and-esg/circularity">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-a0a97faab9-cta-61edadb49a" aria-label="Read more" href="/sustainability-and-esg/communities">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-68c6e69072-cta-c0161c8c65" aria-label="Read more" href="/sustainability-and-esg/sustainability-governance">Read more</a> <a class="cmp-teaser__action-link " id="batcom-teaser-fe169a5262-cta-a72d52f36f" data-cmp-data-layer="{"batcom-teaser-fe169a5262-cta-a72d52f36f":{"@type":"batcom/base/components/batcom-teaser/v1/batcom-teaser/cta","dc:title":"Read more","xdm:linkURL":"/in… | |||||||||||||||||||||
| MS13 | warning | 1 of 1 iframes untitled | |||||||||||||||||||
What we check Each embedded panel on the page carries a title saying what it contains. What good looks like Each embedded panel carries a title saying what it contains. Why it matters The embedded region stays unnamed, so what it holds is left unstated in the page around it. What we found on this page 1 of 1 iframes untitled The evidence, taken from the page as served The iframes with no title attribute, as served
<iframe src="https://www.googletagmanager.com/ns.html?id=GTM-TMH7W3WX" height="0" width="0" style="display:none;visibility:hidden"/> | |||||||||||||||||||||
| DS28 | fail | one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view | |||||||||||||||||||
What we check For a single page, the share of its links to OTHER sites, sampled, that lead nowhere. The site's own links are assessed separately. What good looks like The page's references to other sites open for any reader. Where one refuses or has gone, the page carries the fact itself, shows the reference as plain text rather than as a link, or points to a source that answers. Why it matters The outside sources this page offers as support did not answer when they were asked for, so what the page claims is left resting on material that does not come back. In a regulated setting, where a link is carrying a figure or a required statement, that is the difference between a claim that is evidenced and one that is merely asserted. What we found on this page one link to sources on other sites was refused when it was checked. The page is still published, but the other site turns away requests that do not come from a person using a browser. It is listed in the view The evidence, taken from the page as served The external links checked on this page, with the status each returned
A link that did not open appears here in two forms. A DEAD link points at a page that has gone. A REFUSED link points at a page that is still published, but the other site turns away requests that do not come from a person using a browser, so our request was declined. Both land in the same place: the material this page offers as its support did not come back when it was asked for. The consequence is not a matter of tidiness. Where a link carries the weight of a claim, a figure, a policy or a regulatory statement, and the target does not open, the support behind that claim is missing at the moment it is needed. Three things follow, and each of them costs the organisation. The same fact gets taken from somewhere else, which hands a third party control of how the claim is worded. Or the claim is treated as unsupported and set aside. Or an answer is drawn from what remains on the page, and a partial account is presented as a complete one. A refused link is the worst of the three, because the page still reads as though the evidence sits one step away, so the gap is invisible and the answer carries false confidence. A refusal by another site is still this site's matter to resolve, because this site chose to rest a claim on it. Three practical ways close it. Carry the fact being relied on in the page's own words, so the claim stands without the visit. Keep the reference but stop making it a link, showing the address as plain text, so nothing promises a route that does not open. Or point to a source that answers any reader. Where the reference has to stay live, say beside it what a reader will meet, so no one is sent to a closed door without warning.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the page's references to other sites open for any reader. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 15 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| CT34 | fail | the page's own words are thin: 588 of its 652 words are shared navigation and footer, so it reads as 89% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view | |||||||||||||||||||
What we check Whether a page's own words are so few against a large shared navigation and footer that it reads as near-identical to other pages. What good looks like A page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. Why it matters When shared furniture is most of the text, distinct pages look like copies of each other and the little real content is easily passed over. What we found on this page the page's own words are thin: 588 of its 652 words are shared navigation and footer, so it reads as 89% identical to another page, which a machine can then take as a duplicate and pass over, missing the little content that is its own; the matching page is named in the view The evidence, taken from the page as served This page's own words, lost in shared navigation and footer
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: a page carries enough of its own words that a shared navigation and footer do not make it read as a copy of other pages. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 1 out of a possible 5, because it is one of the supporting checks, so it carries the lightest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
PDF documents 5 documents
The first documents discovered on the estate, scored separately from pages.
▸https://www.bat.com/content/dam/batcom/global/main-nav/investors-and-reporting/reporting/combined-annual-and-sustainability-report/BAT_Annual_Report_2025.pdf 56.5% Likely 4 out of 29
| Document information | |
|---|---|
| Title | BAT Combined Annual and Sustainability Report 2025 |
| Author | British American Tobacco plc (BAT) |
| PDF version, pages, size | 1.7 · 406 pages · 13346 KB |
| Found on | https://www.bat.com/media |
| Check | Result | Evidence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PD19 | warning | the document is large (13346 KB); very large files risk being only partly read, so later sections may be missed | |||||||||||
What we check The size of the file, and whether it passes the point where large documents risk partial processing. What good looks like The file sits within the size where documents process in full. Why it matters A very large file is at risk of being handled only in part, leaving later material out of what comes back. What we found on this page the document is large (13346 KB); very large files risk being only partly read, so later sections may be missed The evidence, taken from the page as served How large this document is
Measured from the file as the site served it.
| |||||||||||||
| PD20 | warning | the document is 406 pages long; later sections of very long documents risk being only partly read | |||||||||||
What we check The number of pages, and whether the document is long enough for its later sections to be at risk. What good looks like The document is short enough that its later pages process in full. Why it matters In a very long document the later chapters risk being left out, so what gets quoted comes only from the front. What we found on this page the document is 406 pages long; later sections of very long documents risk being only partly read The evidence, taken from the page as served How long this document is
Counted from the file as the site served it.
| |||||||||||||
| PD11 | fail | no headings come through in the first page's text, so it reads as one undivided block with nothing to mark where its sections begin | |||||||||||
What we check Whether headings on the document's first page are still recognisable once the text is pulled out. What good looks like Headings on the first page stay recognisable once the text is pulled out. Why it matters With no headings to divide it the document reads as one undivided block, so a passage loses the section it sat in. What we found on this page no headings come through in the first page's text, so it reads as one undivided block with nothing to mark where its sections begin The evidence, taken from the page as served The first page's text as it reads out, with no headings to be found
Combined Annual and Sustainability Report 2025 Building a Smokeles s World Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: headings on the first page stay recognisable once the text is pulled out. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it does not move the misinformation reading This failure makes content harder to reach or to read, but it is not a direct route to a wrong answer, so it is held out of the misinformation reading rather than counted twice. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||
| PD28 | warning | 1 of 9 fonts in the document do not record which character each shape stands for, so some of the text may come out garbled | |||||||||||
What we check Fonts that carry no map turning their shapes back into actual letters, across the sampled pages. What good looks like The document's fonts carry the map turning their shapes back into letters. Why it matters Without that map the text comes out as the wrong characters, so words and figures are wrong rather than missing. What we found on this page 1 of 9 fonts in the document do not record which character each shape stands for, so some of the text may come out garbled The evidence, taken from the page as served The fonts whose text has no character map, and where each is used
A font stores shapes. The map from those shapes back to real characters is what lets the words be read out of the file. Without it the shapes still print correctly on paper, so the document looks perfect, while the text taken from it comes out as the wrong characters. Carrying the font inside the document and recording that map are two separate things: a font can be carried in full and still record no map, and that is the more serious case, because nothing looks wrong at any point and the wrong characters appear only once the document is read by machine. Each font below states which of the two applies to it.
| |||||||||||||
▸https://www.bat.com/content/dam/batcom/global/main-nav/investors-and-reporting/results-centre/pdf/FY_2025_Announcement.pdf 33.7% Very likely 3 out of 29
| Document information | |
|---|---|
| Title | Preliminary results for the year ended 31 December 2025 |
| Author | British American Tobacco |
| PDF version, pages, size | 1.4 · 58 pages · 1224 KB |
| Found on | https://www.bat.com/media |
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PD13 | warning | the document carries no underlying tags marking its headings, figures and reading order, so nothing in the file records how it is meant to be read | |||||||||||||||||||
What we check Whether the document carries the underlying tags that mark up its headings, figures and reading order. What good looks like The document carries an internal structure map, so it is tagged for reading order. Why it matters Without those tags the file records only where things sit on the page, which is why its headings and tables come apart when the text is read out. What we found on this page the document carries no underlying tags marking its headings, figures and reading order, so nothing in the file records how it is meant to be read The evidence, taken from the page as served This document's properties, as recorded in the file
The file was read for the underlying tags that mark up its headings, figures and reading order.
| |||||||||||||||||||||
| PD12 | fail | the document has table-like content, but its column boundaries are lost when the text is read out (31 place(s) where separate cells ran together), so the numbers no longer line up with their row and column labels | |||||||||||||||||||
What we check Whether numbers in tables stay attached to their row and column once the text is pulled out. What good looks like Numbers in tables stay attached to their row and column once the text is pulled out. Why it matters A figure that comes loose from its heading is read against the wrong row or the wrong column, so the number an AI system quotes is not the number the document gives. What we found on this page the document has table-like content, but its column boundaries are lost when the text is read out (31 place(s) where separate cells ran together), so the numbers no longer line up with their row and column labels The evidence, taken from the page as served The text as it reads out, at the places where a value and its label ran together
11.5 9.3 4.8 U.S. AME APMEA 6.53.0 1.7 U.S. AME APMEA Adj Profit from Operations1, 3 (£bn) YoY movements at constant rates -18 % +6 % +10% Reported diluted Adjusted diluted Adj for Canada diluted Adj for Canada diluted at CC 250 300 350 400 Earnings per Share (pence) 349.1 352.1 340.5 5 1 352.8 Change in Group diluted EPS: Reported +157 % Adj1 at CC2 +0.7 % As adj for Canada1,3 -0.2 % As adj for Canada1,3 at CC2 +3.4 % 12 February 2026 – Press Release/Preliminary Results British Amer Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: numbers in tables stay attached to their row and column once the text is pulled out. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| PD14 | warning | the document has images but carries no text descriptions for them, so any figures that live only inside a chart or diagram are absent from the text | |||||||||||||||||||
What we check Whether charts and images in the document carry a written description of what they show. What good looks like Charts and images in the document carry a written description of what they show. Why it matters Data living only inside a chart picture is absent from the document's text, so the figures behind it go unstated. What we found on this page the document has images but carries no text descriptions for them, so any figures that live only inside a chart or diagram are absent from the text The evidence, taken from the page as served Which pages carry images with no text description
Images with no text description appear on page(s): 1, 2, 3, 4, 5, 6, 7, 8 | |||||||||||||||||||||
| PD25 | alert | the same content is not published as an ordinary web page, only as this PDF, which is a harder format to read reliably. This is an alert, not a fault in the document. | |||||||||||||||||||
What we check Whether the same content is also published as an ordinary web page, not only as a download. What good looks like The same content is also published as an ordinary web page, not only as a download. Why it matters Without a page version the content exists only in document form, where its structure is far more fragile. What we found on this page the same content is not published as an ordinary web page, only as this PDF, which is a harder format to read reliably. This is an alert, not a fault in the document. The evidence, taken from the page as served Where an ordinary web page of this content was looked for
Each address below was asked for. A companion page is also matched by a title matching the document's own.
| |||||||||||||||||||||
▸https://www.bat.com/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/BATrsquos-approach-to-corporate-taxation-107-kb.pdf 57.4% Likely 2 out of 29
| Document information | |
|---|---|
| Title | Our approach to corporate taxation 2023 |
| Author | British American Tobacco |
| PDF version, pages, size | 1.6 · 2 pages · 66 KB |
| Found on | https://www.bat.com/who-we-are/our-corporate-governance |
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PD26 | warning | the document is over 2 years old (last dated 2024-01); ageing content is a weaker, less current source | |||||||||||||||||||
What we check How old the document is, measured from the most recent date recorded inside the file. What good looks like The document's most recent internal date is recent rather than long past. Why it matters A document years past its last update still reads as a current statement, because nothing on it says otherwise. What we found on this page the document is over 2 years old (last dated 2024-01); ageing content is a weaker, less current source The evidence, taken from the page as served This document's properties, as recorded in the file
The date shown is the date the file records for itself.
| |||||||||||||||||||||
| PD11 | fail | no headings come through in the first page's text, so it reads as one undivided block with nothing to mark where its sections begin | |||||||||||||||||||
What we check Whether headings on the document's first page are still recognisable once the text is pulled out. What good looks like Headings on the first page stay recognisable once the text is pulled out. Why it matters With no headings to divide it the document reads as one undivided block, so a passage loses the section it sat in. What we found on this page no headings come through in the first page's text, so it reads as one undivided block with nothing to mark where its sections begin The evidence, taken from the page as served The first page's text as it reads out, with no headings to be found
Our approach to corporate taxation We are committed to fully comply with all taxation laws in jurisdictions where we operate, and to account for and pay all taxes that are properly due. Our Tax Strategy, applicable to all our Group companies, is published in our Annual Report and Form 20F. The Tax Strategy is reviewed and approved by the Group Finance and Transformation Director. The operation of the strategy is managed by the Group Finance and Transformation Director and Group Head of Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: headings on the first page stay recognisable once the text is pulled out. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it does not move the misinformation reading This failure makes content harder to reach or to read, but it is not a direct route to a wrong answer, so it is held out of the misinformation reading rather than counted twice. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| PD25 | alert | the same content is not published as an ordinary web page, only as this PDF, which is a harder format to read reliably. This is an alert, not a fault in the document. | |||||||||||||||||||
What we check Whether the same content is also published as an ordinary web page, not only as a download. What good looks like The same content is also published as an ordinary web page, not only as a download. Why it matters Without a page version the content exists only in document form, where its structure is far more fragile. What we found on this page the same content is not published as an ordinary web page, only as this PDF, which is a harder format to read reliably. This is an alert, not a fault in the document. The evidence, taken from the page as served Where an ordinary web page of this content was looked for
Each address below was asked for. A companion page is also matched by a title matching the document's own.
| |||||||||||||||||||||
▸https://www.bat.com/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/BAT_-_2025_Report_on_compliance_with_the_UK_Corporate_Governance_Code.pdf 9.6% Extremely likely 4 out of 28
| Document information | |
|---|---|
| Title | Report on compliance with the UK Corporate Governance Code |
| Author | British American Tobacco |
| PDF version, pages, size | 1.7 · 16 pages · 377 KB |
| Found on | https://www.bat.com/who-we-are/our-corporate-governance |
| Check | Result | Evidence | ||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PD11 | fail | no headings come through in the first page's text, so it reads as one undivided block with nothing to mark where its sections begin | ||||||||||||||||||||||||||||||||||||||||||||||
What we check Whether headings on the document's first page are still recognisable once the text is pulled out. What good looks like Headings on the first page stay recognisable once the text is pulled out. Why it matters With no headings to divide it the document reads as one undivided block, so a passage loses the section it sat in. What we found on this page no headings come through in the first page's text, so it reads as one undivided block with nothing to mark where its sections begin The evidence, taken from the page as served The first page's text as it reads out, with no headings to be found
British American Tobacco p.l.c. Report on compliance with the UK Corporate Governance Code 2024 11 February 2026 Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: headings on the first page stay recognisable once the text is pulled out. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it does not move the misinformation reading This failure makes content harder to reach or to read, but it is not a direct route to a wrong answer, so it is held out of the misinformation reading rather than counted twice. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | ||||||||||||||||||||||||||||||||||||||||||||||||
| PD14 | fail | 1 of 1 image(s) carry no text description of what they show, so figures that live only inside a chart or diagram are absent from the text and go unstated | ||||||||||||||||||||||||||||||||||||||||||||||
What we check Whether charts and images in the document carry a written description of what they show. What good looks like Charts and images in the document carry a written description of what they show. Why it matters Data living only inside a chart picture is absent from the document's text, so the figures behind it go unstated. What we found on this page 1 of 1 image(s) carry no text description of what they show, so figures that live only inside a chart or diagram are absent from the text and go unstated The evidence, taken from the page as served Which pages carry images with no text description
Images with no text description appear on page(s): 1 ; 1 image(s) with no description Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: charts and images in the document carry a written description of what they show. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | ||||||||||||||||||||||||||||||||||||||||||||||||
| PD16 | fail | the document writes out 1 web address as plain text but saves none of them as real, clickable links, so the address sits in the text with no link behind it to follow to the source | ||||||||||||||||||||||||||||||||||||||||||||||
What we check Whether the document's links are recorded as real links, or just web addresses typed as plain text. What good looks like The document's links are recorded as real links, not web addresses typed as plain text. Why it matters When a source is only typed out as text, there is no link behind it to follow, so the reference cannot be opened or checked. What we found on this page the document writes out 1 web address as plain text but saves none of them as real, clickable links, so the address sits in the text with no link behind it to follow to the source The evidence, taken from the page as served The web addresses written out in the text, with no link behind them
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the document's links are recorded as real links, not web addresses typed as plain text. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it moves the misinformation reading This failure could contribute to a wrong answer in some cases, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | ||||||||||||||||||||||||||||||||||||||||||||||||
| PD28 | fail | 8 of 9 fonts in the document do not record which character each shape stands for, so their text comes out as garbled or wrong characters | ||||||||||||||||||||||||||||||||||||||||||||||
What we check Fonts that carry no map turning their shapes back into actual letters, across the sampled pages. What good looks like The document's fonts carry the map turning their shapes back into letters. Why it matters Without that map the text comes out as the wrong characters, so words and figures are wrong rather than missing. What we found on this page 8 of 9 fonts in the document do not record which character each shape stands for, so their text comes out as garbled or wrong characters The evidence, taken from the page as served The fonts whose text has no character map, and where each is used
A font stores shapes. The map from those shapes back to real characters is what lets the words be read out of the file. Without it the shapes still print correctly on paper, so the document looks perfect, while the text taken from it comes out as the wrong characters. Carrying the font inside the document and recording that map are two separate things: a font can be carried in full and still record no map, and that is the more serious case, because nothing looks wrong at any point and the wrong characters appear only once the document is read by machine. Each font below states which of the two applies to it.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the document's fonts carry the map turning their shapes back into letters. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 5 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | ||||||||||||||||||||||||||||||||||||||||||||||||
| PD25 | alert | the same content is not published as an ordinary web page, only as this PDF, which is a harder format to read reliably. This is an alert, not a fault in the document. | ||||||||||||||||||||||||||||||||||||||||||||||
What we check Whether the same content is also published as an ordinary web page, not only as a download. What good looks like The same content is also published as an ordinary web page, not only as a download. Why it matters Without a page version the content exists only in document form, where its structure is far more fragile. What we found on this page the same content is not published as an ordinary web page, only as this PDF, which is a harder format to read reliably. This is an alert, not a fault in the document. The evidence, taken from the page as served Where an ordinary web page of this content was looked for
Each address below was asked for. A companion page is also matched by a title matching the document's own.
| ||||||||||||||||||||||||||||||||||||||||||||||||
▸https://www.bat.com/content/dam/batcom/global/main-nav/who-we-are/our-corporate-governance-/the-board/Holly-430(2B).pdf 30.9% Very likely 3 out of 28
| Document information | |
|---|---|
| Title | Resignation of Holly Keller Koeppel: Disclosures required by section 430(2B), Companies Act 2006 |
| Author | British American Tobacco |
| PDF version, pages, size | 1.7 · 1 pages · 204 KB |
| Found on | https://www.bat.com/who-we-are/our-corporate-governance/the-board |
| Check | Result | Evidence | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| PD08 | warning | filename "Holly-430(2B).pdf" is one word: it hints at the subject but does not identify the document on its own. The file's own title, "Resignation of Holly Keller Koeppel: Disclosures required by section 430(2B), Co", completes it. | |||||||||||||||||||
What we check The file's name, read together with the title in the file's own properties: whether its words, dates and period markers say what the document is. What good looks like A name whose words and dates identify the document, such as FY_2025_Announcement.pdf, passes. A WARNING means the name only half-identifies it: a single hinting word, period markers without a subject (FY25_Q3.pdf), or a weak name that the file's own title completes. A FAIL means a code or generic name, such as IMG_1234.pdf, with no title in the properties either, so nothing anywhere says what the document is. Why it matters When the name says nothing and the properties carry no title, the document has no usable name anywhere, so a reader, human or AI, cannot tell what it is before opening it, and it is cited or skipped blind. What we found on this page filename "Holly-430(2B).pdf" is one word: it hints at the subject but does not identify the document on its own. The file's own title, "Resignation of Holly Keller Koeppel: Disclosures required by section 430(2B), Co", completes it. The evidence, taken from the page as served This document's properties, as recorded in the file
The file name and the file's own properties are read together: a title in the properties can complete a name that is weak on its own.
| |||||||||||||||||||||
| PD11 | fail | no headings come through in the first page's text, so it reads as one undivided block with nothing to mark where its sections begin | |||||||||||||||||||
What we check Whether headings on the document's first page are still recognisable once the text is pulled out. What good looks like Headings on the first page stay recognisable once the text is pulled out. Why it matters With no headings to divide it the document reads as one undivided block, so a passage loses the section it sat in. What we found on this page no headings come through in the first page's text, so it reads as one undivided block with nothing to mark where its sections begin The evidence, taken from the page as served The first page's text as it reads out, with no headings to be found
BRITISH AMERICAN TOBACCO p.l.c. (the “Company”) Resignation of Holly Keller Koeppel: Disclosures required by section 430(2B), Companies Act 2006 As required by section 430(2B) of the Companies Act 2006, the Company confirms that no remuneration is payable consequent upon or in respect of her resignation. Holly Keller Koeppel received pro-rated Board and Nomination Committee fees to the date of her resignation. She will not receive any payment for loss of office or any other payment in rela Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: headings on the first page stay recognisable once the text is pulled out. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 3 out of a possible 5, because it is one of the checks that decide how cleanly content can be read once reached, so it carries a middle weight in the figure for this item. Why it does not move the misinformation reading This failure makes content harder to reach or to read, but it is not a direct route to a wrong answer, so it is held out of the misinformation reading rather than counted twice. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| PD28 | fail | 2 of 2 fonts in the document do not record which character each shape stands for, so their text comes out as garbled or wrong characters | |||||||||||||||||||
What we check Fonts that carry no map turning their shapes back into actual letters, across the sampled pages. What good looks like The document's fonts carry the map turning their shapes back into letters. Why it matters Without that map the text comes out as the wrong characters, so words and figures are wrong rather than missing. What we found on this page 2 of 2 fonts in the document do not record which character each shape stands for, so their text comes out as garbled or wrong characters The evidence, taken from the page as served The fonts whose text has no character map, and where each is used
A font stores shapes. The map from those shapes back to real characters is what lets the words be read out of the file. Without it the shapes still print correctly on paper, so the document looks perfect, while the text taken from it comes out as the wrong characters. Carrying the font inside the document and recording that map are two separate things: a font can be carried in full and still record no map, and that is the more serious case, because nothing looks wrong at any point and the wrong characters appear only once the document is read by machine. Each font below states which of the two applies to it.
Why it reads as a failure The result is set by what was measured on this item, quoted above, and not by an opinion of it. It reads as a failure because the standard was not met: the document's fonts carry the map turning their shapes back into letters. What it does to the figure This check counts toward the readiness figure for this item, at a weight of 5 out of a possible 5, because it is one of the checks that decide whether content can be reached and read at all, so it carries the heaviest weight in the figure for this item. Why it moves the misinformation reading This failure is a real and common route to a wrong answer for the content it affects, so it raises the misinformation reading for this item. Where an item carries more than one such failure, the readings compound and the item moves up the scale. What this result does not say This says what was found on this item at the time of assessment, and nothing beyond it: it is not a statement about the rest of the estate, and it does not judge the quality of the content itself. | |||||||||||||||||||||
| PD25 | alert | the same content is not published as an ordinary web page, only as this PDF, which is a harder format to read reliably. This is an alert, not a fault in the document. | |||||||||||||||||||
What we check Whether the same content is also published as an ordinary web page, not only as a download. What good looks like The same content is also published as an ordinary web page, not only as a download. Why it matters Without a page version the content exists only in document form, where its structure is far more fragile. What we found on this page the same content is not published as an ordinary web page, only as this PDF, which is a harder format to read reliably. This is an alert, not a fault in the document. The evidence, taken from the page as served Where an ordinary web page of this content was looked for
Each address below was asked for. A companion page is also matched by a title matching the document's own.
| |||||||||||||||||||||
How the scores work method & bands
This report is Stage 1, Foundations: each automated test the engine can run without human input, so there is a usable result at the first stage. Governance (owner-confirmed truth) and Visibility (how AI agents find the estate) are later stages and are shown greyed above.
What was assessed. 37 pages were assessed, selected from 120 reviewed using the site's own signals of importance (its navigation, homepage links, most-linked pages, sitemap, and any llms.txt file), and the first 5 PDF documents found were assessed separately. The industry in the header is classified from the site's own content. The traffic classification comes from a recognised top-sites list or a connected traffic provider; it reads NA when the domain is not connected to a traffic source.
Header identity signals. The regulated sector and listed company values are inferred from the same crawled content, and shown here so the header can stay to the answer alone. Regulated sector reads “Yes”: tobacco and nicotine sector; no explicit regulator statement in the sampled pages. Listed company reads “Yes (public)”: traded on london stock exchange (bats), new york stock exchange (adr) (bti).
AI readiness is the weighted share of checks that passed, reported separately for pages and for PDF documents because each is reached and corrected differently. Grade A checks decide whether content can be reached and read at all; grade B whether it is understood correctly; grade C support reliability. A measured score is scaled into a ceiling when an access gate binds it (for example the site blocking AI crawlers at its edge): content that cannot be reached is not partially ready. Scores never read exactly 25, 50, 75 or 100, because those are buckets, not measurements.
Likelihood of misinformation is a band, given as a worst case (the highest any single item or estate-wide finding reaches) and a typical position across the pages or documents. Only checks whose failure is a direct route to a wrong answer carry a band; the rest affect readiness only.
| Band | Range | Meaning |
|---|---|---|
| Likely zero | 0-5% | No failing misinformation-bearing checks. |
| Very little | 6-25% | Unlikely to cause wrong information. |
| Likely | 26-50% | Could contribute to wrong information in some cases. |
| Very likely | 51-75% | A real and common route to wrong information. |
| Extremely likely | 76-100% | A frequent and direct cause of wrong information. |
Of the register's 187 checks, 113 ran automatically in this Foundations stage. 64 are governance questions for the next stage. The remainder need organisation-supplied data or an external service; each check that did not run is listed in the exports.
About us
Built on evidence, not assumptions.
We have been doing this for 25 years. For the last two years, we have been training AI to understand and map AI Readiness (fundamentals, governance, visibility), against our unique dataset of more than 3.7 trillion data points.
3.7
Trillion
Data points, with each result kept since the beginning: probably the largest experience and compliance dataset of its kind. Our AI is trained on factual, controlled, real-world evidence.
$106
Billion
Estimated manual effort avoided through automation. An estimate, and a conservative one: since ARTY was introduced, we have removed 95% of costs and delivered efficiency gains beyond 70x.
119
Million
Websites analysed and mapped for content fundamentals, from the outside in. AI Readiness begins by removing the human issues that undermine Value and create unnecessary Risk.
25
Years
When we started, we were labelled 'the pioneers' of website automation and checking. In 2023, we launched our first AI analysis tools.
Who’s talking about us
