BlogFindings

AboutPage schema: 31 of 623 about pages declared it, and none declared the type Google documents

Lantad asked all 1,419 hostnames in this repository's two committed corpus seed files for robots.txt on 18 October 2026, read every permitted home page with no JavaScript executed, then followed each page's own link to an about, company or team surface and read that too. 623 of those pages answered HTTP 200 with HTML. 31 declared schema.org AboutPage, a type that defines no properties of its own, and not one declared ProfilePage, the page type Google's structured data gallery documents for a page about a single person or organisation.

21 min read Lantad

The answer separates into two findings that point in different directions. The first is that AboutPage markup is close to absent and would not buy much if it were present, because the type carries no properties and no Google feature reads it. The second is the one worth acting on: on 100 of the 623 sites measured here, the organisation node that the home page declares is simply gone by the time a crawler reaches the about page, and on 41 of them the page delivered under 100 words. The markup gets thinner one click in, and so does the prose. Lantad ran this on 18 October 2026 across the 1,419 hostnames described on the methodology page, reading one about surface per site as LantadBot with no browser involved.

In short

  • AboutPage schema is a label rather than a carrier: schema.org release 30.1, dated 2026-09-16, defines the type as "Web page type: About page" and gives it no properties of its own, and Lantad found it declared on 31 of the 623 about pages it read on 18 October 2026.
  • Not one of those 623 pages declared ProfilePage, which is the page type Google's structured data gallery documents for a page whose subject is a single person or organisation, in a gallery last updated on 2026-10-16 UTC.
  • Machine-readable identity thins out one click past the front door: 420 of the 623 sites carried parseable JSON-LD on their home page against 340 on their about page, with 95 dropping it entirely and 15 gaining it.
  • 100 of the 623 sites declared an organisation node on their home page and none at all on the about page that page itself linked, measured by Lantad on 18 October 2026.
  • Of the 249 about pages that did declare an organisation, 157 declared none of founder, foundingDate or address, the three facts an about page exists to state.
What the about page declaredOf the 623What it is
An organisation node249Organization or one of its subtypes, with or without properties
A WebPage node with no subtype121The generic page type, which says nothing about the subject
A Person node57151 nodes between them, every one carrying a name
AboutPage31A schema.org type that defines no properties of its own
mainEntity anywhere in the markup28The property Google's ProfilePage documentation requires
ProfilePage0The page type Google's gallery documents for this job
No JSON-LD at all283No typed node of any kind in the delivered bytes
What the 623 about pages declared, read once each by Lantad on 18 October 2026 as LantadBot/1.0 with no JavaScript executed, every JSON-LD block flattened with flattenDeep and typed with entityTypes in core/src/jsonld.ts. 619 of the 623 answered a second request the same day and 618 of those reproduced all four of the presence figures below.

What is AboutPage schema, and what does declaring it do?

schema.org defines AboutPage in six words: "Web page type: About page." Its place in the vocabulary is Thing, then CreativeWork, then WebPage, then AboutPage, and the page carries version 30.1 dated 2026-09-16 in its footer. The important detail is what is not there. AboutPage defines no properties of its own. Every property it lists, breadcrumb and speakable and specialty among them, is inherited from WebPage or further up. Declaring the type therefore adds a label to a page and no facts about the organisation on it.

That is worth stating plainly because the label is easy to mistake for the work. A site can declare AboutPage, pass every validator, and still have told a reader nothing except that this page is an about page, which the URL usually said already. The facts a reader wants, who founded the organisation and when and where it is, live on an Organization or Person node, and whether those nodes are present is a separate question this post measures separately below.

Google's position is the second half of the frame. Its structured data gallery, last updated 2026-10-16 UTC, documents 25 features that structured data can produce in Search. Article, Breadcrumb, Organization, Product, Profile page and Q&A are on that list. AboutPage is not. There is no documented Search feature that reads it, which means a site declaring it is not buying a rich result and is not failing to buy one either. For the AI surfaces specifically, Google's AI features documentation, last updated 2025-12-10 UTC, is blunter still: "There's also no special schema.org structured data that you need to add," and eligibility to appear as a supporting link requires only that a page be indexed and eligible to be shown with a snippet, with "no additional technical requirements."

So the honest expectation going in was that AboutPage would be rare and that its rarity would cost nobody anything. 31 of 623 is rare. The rest of this post is about the things on the same pages that are not free to omit, and the pattern that shows up when the about page is compared with the home page of the same site. This is the same measurement discipline applied to per-type schema counts across the corpus, moved one click inward.

Where AboutPage sits in the schema.org vocabulary, read from schema.org/AboutPage on 18 October 2026, which carries version 30.1 dated 2026-09-16. The type inherits every property it lists and defines none.

How a crawler reaches an about page at all

The corpus is fixed and committed: 1,419 hostnames across two seed files, 392 in ten platform strata and 1,027 in eight industry strata. Each was asked for /robots.txt once on 18 October 2026, as LantadBot/1.0 with redirects followed and a twenty second timeout. 29 requests failed before any status arrived, 1,055 files parsed into at least one user-agent group, 292 answered a status other than 200 and 43 answered 200 with a content type that was not text/plain. Every root verdict was computed with this scanner's own evaluateRobots in core/src/robots.ts, which left 1,378 hostnames permitting LantadBot at the site root and 12 refusing it.

Those 1,378 home pages were then requested once each. 1,077 answered HTTP 200 with HTML, 299 answered another status and two failed in the client. From each of those 1,077 pages, every same-host anchor was scored and the shallowest one whose URL path or anchor text explicitly said it was an about, company, story, team, people, leadership or mission surface was taken. 632 of the 1,077 produced such a link, 544 of them matched on the path and 79 on the anchor text alone. 307 pages linked no about surface a crawler could follow on their own host, and a further 138 linked one only on a different hostname, which is a different measurement and was left out rather than mixed in.

The 632 picked URLs were each checked against the same site's parsed robots.txt before being requested, and none was disallowed. 623 answered HTTP 200 with HTML, six answered 403 and three answered 404. Every page was then read a second time the same day. 619 answered again, and 618 of those reproduced the presence or absence of parseable JSON-LD, of AboutPage, of a Person node and of an organisation node exactly. The single change was gov.scot, which carried no parseable block on the first read and one on the second. Four hosts answered a different status second time around, those being bangkokpost.com with a 451, plannedparenthood.org with a 403, revolve.com with a 503 and time.com with a 406.

One picked link in this set is worth naming, because the selector is mechanical and sometimes lands beside the target rather than on it. atlassian.com's shallowest matching link went to /company/contact/general-inquiries, ca.gov's to /about/sitemap and inquirer.com's to /about/copyright. Those are about-section pages rather than about pages, and they are counted as what they are: the surface the home page advertised first under an about-shaped label. The selector is also English-language, which systematically undercounts sites that label the same page in another language, and that limit applies to a corpus holding a large number of non-English government and news hosts. An earlier run over this corpus on 6 October 2026 matched a looser word list and reached 709 about pages where this one reaches 623, so the two sets overlap heavily without being identical and the counts are not interchangeable. The equivalent funnel for other surfaces, such as 220 documentation and help pages and 195 pricing pages, was built the same way.

StepCountWhat happened to the rest
Hostnames in the two seed files1,419392 platform strata, 1,027 industry strata
Permitted LantadBot at the site root1,37812 refused it, 29 failed before any status
Home pages answering 200 with HTML1,077299 answered another status, 2 failed in the client
Linked a same-host about surface632307 linked none, 138 linked one off-host only
About pages answering 200 with HTML6236 answered 403, 3 answered 404
Reproduced on a second read the same day618 of 619gov.scot gained a parseable block, 4 changed status
From the committed corpus to the 623 about pages, measured by Lantad on 18 October 2026. Robots verdicts computed with this scanner's parseRobotsTxt and evaluateRobots in core/src/robots.ts, never observed from crawler traffic.

The type Google documents is ProfilePage, and none of the 623 declared it

There is a page type in Google's gallery for exactly this job, and it is not AboutPage. Google's profile page documentation, last updated 2026-09-08 UTC, says ProfilePage markup "is designed for any site where creators (either people or organizations) share first-hand perspectives," and its own list of examples names author pages on news sites, "About Me" pages on blogs and employee pages on company websites. It carries a restriction worth quoting, which is that "the primary focus of the page must be a single person or organization that is affiliated with the overall website," and it names two required properties: mainEntity, holding the Person or Organization the page is about, and a name on that entity.

Not one of the 623 pages declared ProfilePage. That is a clean zero in a corpus where 249 of the same pages declared an organisation and 57 declared a person, so the entities ProfilePage asks for were frequently on the page already, unlabelled as its subject. The property that does the labelling was nearly as scarce: 28 of the 623 used mainEntity anywhere in their markup, and 28 used the related about property. Of the 31 pages that did declare AboutPage, 15 attached a mainEntity to it and 16 declared the type with nothing named as the page's subject.

The gap is not quite a recommendation, and this is where the honest reading matters. ProfilePage is documented in the context of the Discussions and Forums feature and of creators publishing first-hand perspectives, so a corporate about page describing a company is not the central case its documentation addresses, and declaring it would not make a company page eligible for a feature built around community content. What the zero does establish is that the one page type Google documents for a page whose subject is a single named entity is unused across this corpus, while the type that is used instead carries no properties and no documented feature. Sites are reaching for the label the vocabulary offers rather than the one the search engine reads.

The 31 that declared AboutPage are a short enough list to name, and naming them is the point of a measurement anybody can check: agentplace.io, aiwyn.ai, alliantcreditunion.org, amplemarket.com, anrok.com, appalshop.org, asana.com, bdmasonry.net, bitrise.io, carecycle.ai, crayo.ai, edexia.com, embl.org, emiratesnbd.com, fin.ai, highlandscurrent.org, hookagency.com, povio.com, principal.com, runcopycat.com, spekit.com, statnews.com, techwolf.ai, thewireguyelectric.com, trustmrr.com, typeform.com, uipath.com, uplisting.io, weareelectrical.co.uk, webnames.ca and whipsaw.com. By stratum the type clusters in software and in builder-generated sites, with seven of 37 Webflow sites and six of 85 SaaS sites declaring it, against none of 55 healthcare sites, none of 45 Wix and Squarespace sites and none of 30 government sites. The same clustering by vertical turned up in the corpus measurement of organisation subtypes.

  • Declared ProfilePage 0 of 623 The type Google documents for a page about one named entity
  • Declared AboutPage 31 of 623 A type with no properties and no documented Search feature
  • Declared mainEntity anywhere 28 of 623 One of the two properties ProfilePage requires
  • AboutPage carrying a mainEntity 15 of 31 16 declared the type and named no subject for the page
  • Had an entity available to name 265 of 623 249 declared an organisation, 57 a person, with overlap
  • Google feature reading AboutPage None documented 25 features listed in the gallery, AboutPage not among them
ProfilePage against AboutPage on the 623 about pages read by Lantad on 18 October 2026. Documentation facts read at source the same day from developers.google.com and schema.org; the counts are of declarations in delivered HTML, not of any engine's behaviour.

Structured data thinned out one click past the home page

Because the home page of every one of these 623 sites was read in the same run, the two surfaces can be compared directly, and that comparison is the finding with the most consequence in this post. 420 of the 623 sites carried at least one parseable JSON-LD block on their home page. 340 carried one on their about page. 95 sites carried markup at the front door and none at all one click in, and 15 did the reverse. Every page that carried a block had at least one that parsed, so this is a presence gap rather than a syntax problem of the kind measured in 117 of 612 home pages with a JSON-LD defect.

The organisation node follows the same slope and more steeply. 341 of the 623 declared an organisation node on the home page and 249 did on the about page, so 100 sites state in machine-readable form who they are on the page that sells the product and say nothing on the page that answers the question. aljazeera.com, aiven.io, atlassian.com, ally.com, blackrock.com, capitalone.com, browserstack.com, aafp.org, ama-assn.org, bostonglobe.com and becu.org are among them, and the list is not weighted toward small sites that never got round to it. A crawler that follows the about link from any of those home pages receives prose and no typed identity.

Person nodes are the exception, and only because they were already scarce. This blog measured them on these same two surfaces on 6 October 2026 and published the result: 111 of 1,079 sites named a human, 90 of them on a home page and 71 on an about page, with the sector detail and the property gaps set out there. This run reproduces that shape on a stricter funnel, finding a Person node on 68 of the 623 home pages and 57 of the 623 about pages, 26 on the home page only and 15 on the about page only. The person figures here are therefore a replication rather than a new finding. The organisation node is the thing nobody had counted on the about surface, and it is the thing that moves.

Where the markup does survive, it tends to be a builder emitting it rather than an author writing it. The platform strata, which are sites grouped by the tool they are built with, carried parseable JSON-LD on 171 of 264 about pages. The industry strata carried it on 169 of 359. That is 64.8 percent against 47.1 percent, and the WordPress and Wix and Squarespace strata sit at the top of the first group, and the same front-door gap on a smaller frame produced 141 of 382 home pages carrying none in the raw HTML. A template that emits an Organization node on every page is doing the work that nobody on the industry sites did by hand.

  • Parseable JSON-LD, home page 420 sites Of 623 sites
  • Parseable JSON-LD, about page 340 sites 95 dropped it, 15 gained it
  • Organisation node, home page 341 sites
  • Organisation node, about page 249 sites 100 declared one at the front door only
  • Person node, home page 68 sites
  • Person node, about page 57 sites 540 of 623 declared one on neither surface
The same 623 sites, home page against about page, measured by Lantad on 18 October 2026 in one run with no JavaScript executed. Each bar counts sites declaring the thing at least once in delivered HTML.

The facts an about page exists to carry, and how many stated them

Set the page types aside and ask what the 249 organisation nodes on these about pages actually say. 235 of them carry a name. After that the properties drop away fast. 158 carry sameAs, which is the property that points at an external reference entry and the one this blog found 85 of 317 organisations using on home pages. 70 carry an address. 40 carry a foundingDate. 29 name a founder. Seven state a numberOfEmployees.

The combination is the useful number. 157 of the 249 declare none of founder, foundingDate or address, which are the three facts an about page exists to state and the three a reader arriving from an answer engine is most likely to want. 63 of the 249 carry a name and nothing else at all from the set of five properties counted here. A node like that asserts that an organisation exists and is called something, which the title element had already established.

The 29 that name a founder are the ones that read as authored rather than generated, against the 60 of 1,079 home pages that declared founder or founders when this corpus was read on 6 October 2026, and they are checkable: asana.com/company, amplemarket.com/about, article.com/aboutus, bitrise.io/about, confluent.io/about, huel.com/pages/about-us, netlify.com/about, ncl.com/about, nordlayer.com/about-us, vercel.com/about and warbyparker.com/history among them. Seven sites state a headcount, those being amplemarket.com, bitrise.io, huel.com, metalbear.com, nami.org, povio.com and statnews.com. Nothing in Google's documentation requires any of these properties. Its Organization structured data guidance, last updated 2026-09-08 UTC, says in its own words that "there are no required properties; instead, we recommend adding as many properties that are relevant to your organization", so this is not a compliance gap. It is a question of whether a machine reading the page can answer "who is this" without parsing prose.

The reason that matters for AI visibility rather than for rich results is the one this site keeps returning to: an answer engine assembling a sentence about a company is reading whatever it can resolve cheaply, and a typed node naming a founder and a founding date is cheaper to resolve than the same facts in a paragraph. That is an argument about mechanism, not a measured effect, and this post does not claim an effect. What this run can say is that on the page where those facts are most expected, 157 of 249 organisation nodes carry none of them, and 283 of the 623 pages carry no node at all. The site's own entity confidence definition sets out which of these signals it scores and why, and the broader question of what markup moves is covered on the research page.

PropertyOf the 249 nodesOf all 623 pagesWhat it answers
name235235What the organisation is called
sameAs158158Which external entry refers to the same organisation
address7070Where it is
foundingDate4040When it started
founder2929Who started it
numberOfEmployees77How big it is
None of founder, foundingDate or address157157An organisation exists and has a name
Properties on the 249 organisation nodes found on about pages, Lantad, 18 October 2026. Presence was checked and never correctness, so a wrong founding date counts the same as a right one.

Words came before markup on 41 of the 623

Markup is the second question. The first is whether the page says anything at all in bytes a reader without a browser receives, and that was measured with this scanner's own extractFromHtml in core/src/extract.ts, which drops script, style, noscript, template, svg and iframe content before counting. The median about page delivered 446 words, the quartiles sit at 267 and 766, and 96 pages delivered more than 1,000. The largest was dcu.org/about/rates.html at 10,016 words, followed by webnames.ca at 8,727 and hdfc.bank.in at 8,030.

At the other end, 41 of the 623 delivered under 100 words and five delivered none. The five are airbaltic.com/en/cargo/about-us, gov.scot/about, jetbrains.com/company, santafe.edu/about and pto.nz/about. Four of the five shipped a substantial document to do it: santafe.edu/about sent 105,257 bytes of HTML and no extractable words, pto.nz/about 91,689, jetbrains.com/company 44,349 and airbaltic.com 23,602, while gov.scot/about sent 1,983 bytes, which is a redirect-shaped stub rather than a page. 18 of the 41 thin pages shipped more than 100,000 bytes, among them moh.gov.sg/about-us at 420,912 bytes for 85 words, singleparentproject.org/team-1 at 1,032,044 bytes for 99 words and trip.com/pages/about-us at 244,567 bytes for 38 words.

A large document that extracts to nothing is the signature of a page whose prose arrives only when JavaScript runs, which this blog has measured directly: JavaScript supplied 7.6 percent of the prose on 271 pages and all of it on 11 of them. No browser was involved here, by design, so these counts state the floor every reader shares rather than a prediction about what a rendering crawler ends up holding. Google documents that it renders, on a queue it does not bound, and no other AI crawler vendor documents either way.

Structure thinned too, and in a way that matters for extraction rather than for appearance. 24 of the 623 pages carried no heading element at all, against a median of 11, which leaves an answer engine no labelled section to lift from a page that may well be long. That is the same failure counted at corpus scale in 379 of 1,066 home pages declaring no main landmark. The practical version of all of this takes under a minute per site: request your own about URL with no JavaScript and count what comes back, which is what this tool does, and then ask whether the identity node your home page declares is still there.

GET https://santafe.edu/about as LantadBot/1.0

  • GET /robots.txt 200, parsed, root allowed for LantadBot
  • GET / (home page) 200 text/html, about link found at /about
  • GET /about 200 text/html, 105,257 bytes
  • extractFromHtml, word count 0 words
  • JSON-LD blocks that parsed 0
  • GET /about, second read same day 200, 0 words again
One of the five about pages that delivered no extractable words, requested twice by Lantad on 18 October 2026 with identical results. Word count from the shipped extractFromHtml in core/src/extract.ts, which drops script, style, noscript, template, svg and iframe content.

What this measured, and what it did not

One page per site, and that page was the shallowest surface the home page itself linked under an about-shaped label. These counts describe the door a site advertised, not its full identity material. A site whose about page declares nothing may well declare a complete Organization node on a team page, a leadership page or a press page none of which this run requested, and nothing here contradicts that. The selector is mechanical and English-language, so it lands beside the target on sites like atlassian.com and ca.gov and misses the equivalent page entirely on sites that label it in another language.

No JavaScript was executed and no browser was involved. That bounds every count above to the delivered bytes, which is the floor every reader shares and not a claim about what any particular engine finally holds. Presence was checked and never correctness: a foundingDate that is wrong counts here exactly as one that is right, and a Person node naming someone who left the company in 2023 counts as a named person. The schema counts come from flattening every JSON-LD block, so markup expressed in microdata or RDFa is counted as absent, which this blog has measured separately at 22 of 385 home pages carrying microdata.

LantadBot is not GPTBot. The robots verdicts here were computed from parsed files with this scanner's own evaluator rather than observed from crawler traffic, so a site that admitted LantadBot may answer a different crawler differently, and the six 403 responses are the same kind of unknown rather than evidence of a policy anybody set. The reference list of the tokens involved is on the AI crawler page, and the definition this site works from is at AI crawler.

Most of all, nothing here connects any of these declarations to being cited. No answer engine was queried in this run and no citation was counted, so no sentence above claims that declaring AboutPage, or a founder, or anything else, changes whether an engine names a company. What can be said is narrower and still useful: the page a site points readers to when they ask who is behind it carried no typed node at all on 283 of 623 sites, lost the organisation node its own home page declares on 100 of them, and delivered under 100 words on 41. Those are all things a site owner can fix and check in an afternoon, and the platform-specific routes into Google AI Overviews and ChatGPT citations start from a page that can be read at all. The corpus itself is two editorial sampling frames rather than a sample of the web, described in full at AI visibility, so no proportion above generalises past these 1,419 hostnames.

Measured

  • 632 of 1,077 readable home pages linked a same-host about surface
  • 31 of 623 about pages declared AboutPage, 0 declared ProfilePage
  • 283 of 623 carried no JSON-LD at all
  • 100 of 623 lost the organisation node their home page declares
  • 41 of 623 delivered under 100 words, 5 delivered none
  • 618 of 619 second reads reproduced every presence figure

Not measured

  • Whether any engine cited any of these pages
  • What a browser would render on the 41 thin pages
  • Identity material on team, leadership or press pages
  • Whether any declared property is actually correct
  • Whether a real AI crawler was served the same bytes
  • About pages labelled in a language the selector missed
What the 18 October 2026 run establishes and what it leaves open. Both panels describe the same measurement of 623 about pages.

Written by

Lantad

Published .

An about page is the one surface on a site whose entire job is to say who is behind it. That makes it the natural place to look for the thing structured data is supposed to supply: a machine-readable statement of identity, sitting on the page a reader is sent to when they want to know who they are dealing with. schema.org even defines a page type for it. The question this post settles is a narrow one, which is whether the sites in this repository's corpus use it, and what a reader that executes no JavaScript actually receives when it follows the link.

Common questions

Should I add AboutPage schema to my about page?

It will not hurt and it buys nothing documented. schema.org defines AboutPage as "Web page type: About page" and gives it no properties of its own, and Google's structured data gallery, last updated 2026-10-16 UTC, lists 25 features without it. The declaration that carries facts is an Organization or Person node with properties on it: of the 623 about pages Lantad read on 18 October 2026, 249 declared an organisation and 157 of those 249 declared none of founder, foundingDate or address.

What is the difference between AboutPage and ProfilePage?

ProfilePage is the type Google documents, and it requires properties. Its documentation, last updated 2026-09-08 UTC, says it is designed for sites where creators share first-hand perspectives, names About Me pages and employee pages among its examples, requires a mainEntity holding the Person or Organization the page is about, and restricts eligibility to pages whose primary focus is a single entity affiliated with the site. AboutPage requires nothing because it defines nothing. None of the 623 pages measured declared ProfilePage and 28 declared mainEntity at all.

Does structured data on an about page affect AI search visibility?

This measurement does not show that, and no claim above says it does. Google's AI features documentation, last updated 2025-12-10 UTC, states there is no special schema.org structured data needed to appear in AI Overviews or AI Mode and that eligibility requires only being indexed and eligible for a snippet. What Lantad measured is availability rather than effect: on 283 of 623 about pages a crawler received no typed node of any kind, and on 41 it received under 100 words.

Why compare the about page with the home page?

Because the comparison isolates a drop that a single-page scan cannot see. Both surfaces were read in the same run on 18 October 2026 for all 623 sites. 420 declared parseable JSON-LD on the home page against 340 on the about page, 95 dropped it entirely, and 100 declared an organisation node on the home page and none on the about page. A site that assumes its markup is site-wide because its home page passes a validator is frequently wrong one click in.

See what AI can read on your site

Run a free scan and get a graded report of exactly what AI crawlers can and cannot read, with ranked fixes.

Scan your site free How we measure

Related

Glossary

Structured data for AI (schema markup)

Structured data is schema.org markup, usually written as JSON-LD, that labels a page's facts in a machine-readable format so AI systems can extract and attribute them without having to interpret the surrounding prose.

 Glossary

Entity confidence

Entity confidence is a free diagnostic on every graded Lantad report that scores, out of 100, how completely the JSON-LD in your crawler-visible HTML declares one identity and how far that identity is anchored by identifiers which resolve outside the page.

 Article

Person schema: 111 of 1,079 sites named a human, and 162 of those names appeared nowhere else on the page

Lantad asked all 1,419 hostnames in this repository's two committed corpus seed files for robots.txt on 6 October 2026 as LantadBot, read every permitted home page with no JavaScript executed, then followed each page's own link to an about, team or company page and read that too. 1,079 home pages and 709 about pages answered HTTP 200 with HTML. 111 of the 1,079 sites declared a Person node on one of those two pages, 418 declared an organisation and no person at all, and of the 518 named Person nodes found, 272 carried neither of the two properties Google's own documentation names for working out who an author is.

 Article

Organization schema type: 277 of 448 identity nodes declared no category, and 55 of 63 subtypes matched the industry

Lantad requested the home page of each of the 1,419 hostnames in this repository's two committed corpus seed files on 17 October 2026, as LantadBot with no JavaScript executed, and ran the shipped entity assessor over the 1,084 that delivered HTML. 448 declared an identity node. 171 stated what kind of organisation they are and 277 did not. Of the 63 industry-corpus sites that stated it through the type itself, 55 chose a subtype belonging to their own sector.

 Article

Entity SEO: 107 of 163 schema topic statements named the site itself

Lantad asked all 1,419 hostnames in this repository's committed corpus for robots.txt on 8 October 2026 and read each home page it was allowed to read, with no JavaScript executed. 1,074 answered HTTP 200 with HTML and 612 of those carried parseable JSON-LD. 129 of the 612 used about, mainEntity or knowsAbout to name a subject, and of the 163 about and mainEntity values across 104 sites, 107 pointed at the site's own organisation, service or founder rather than at a topic.

 Article

Schema markup types: 88 of the 939 schema.org defines reached 626 home pages, and the top three name the page

Lantad asked all 1,419 hostnames in this repository's two committed corpus seed files for robots.txt on 16 October 2026, then read every permitted home page as LantadBot with no JavaScript executed. 1,085 answered a 2xx with HTML, 631 carried at least one JSON-LD block and 626 held one that parsed. Across the 2,181 typed nodes those pages declared, 88 distinct type strings appeared, against the 939 classes schema.org release 30.1 defines. WebSite, Organization and WebPage accounted for 1,029 of the 2,227 type declarations, and five of the 88 strings are not types schema.org defines at all.