Structured Data on the Top 1,000 Homepages: What AI Crawlers See

Key findings

• 47.5% of top homepages carry JSON-LD in the HTML they serve, and 51.9% carry structured data of any kind, microdata and RDFa included.
• Only 31.3% identify the company behind the site with an Organization or Corporation node. Of those, 85% link to their profiles elsewhere with sameAs.
• WebSite is the most common type, on 29.8% of homepages, and 96 still declare a SearchAction.
• FAQPage is rare on homepages: 13 of 486.
• 22.8% of homepages show fewer than 100 words before JavaScript runs. A crawler that doesn’t render sees an almost empty page. The median homepage has 709 words in its raw HTML.
• The basics are nearly universal: 95% have a title, 87% declare a language, 83% a meta description and 70% a canonical URL.

Why raw HTML is the number that matters

Vercel’s December 2024 analysis of crawler traffic found that “none of the major AI crawlers currently render JavaScript”, including GPTBot, ClaudeBot and PerplexityBot (Vercel). Googlebot renders pages; the AI crawlers, by that account, take what the server sends. So JSON-LD injected by a tag manager, or a homepage built entirely in the browser, can look complete in your browser and empty to them. Everything on this page was measured that way: one request, no JavaScript.

The full picture

Does structured data get you cited?

Nobody has shown that it does. Google says no special markup is needed to appear in AI Overviews or AI Mode (Google Search Central), and the AI companies’ crawler documentation doesn’t mention schema. What an Organization node with sameAs links does is state, in one machine-readable place, who you are and where else you exist, which is the kind of fact assistants get wrong. Our guide on schema markup and AI citations goes through the evidence.

What to check on your own homepage

1. View the source, not the inspector. If your JSON-LD only appears after scripts run, crawlers that don’t render won’t see it.
2. Add an Organization node with your name, URL, logo and sameAs links to your real profiles.
3. Put your main claim in text, in the HTML. One plain sentence on what you sell and to whom, readable without JavaScript.
4. Run the structured data checker, which reads your page the same way this study did.

Method and limits

We took the top 10,000 domains of the Tranco list GQ3LK (a research ranking that combines Chrome UX Report, Cloudflare Radar, Cisco Umbrella, Majestic and Farsight data over September 28 to October 27, 2026). On October 28, 2026 our crawler requested each domain’s homepage over HTTPS, following redirects. Many Tranco entries are CDN, API or tracking domains with no website: 2,155 gave no HTTP answer at all and 1,415 answered 403. We kept the 5,480 domains whose homepage returned an HTML page with status 200, counted once per final hostname after redirects. This part of the study uses the 486 websites ranked in the top 1,000; 3 homepages over 3 MB were left out.

Each homepage was read with the parser behind our structured data checker: JSON-LD blocks, their schema.org types, microdata and RDFa attributes, robots meta tags and the X-Robots-Tag header, and the number of words of visible text in the HTML as served. Types are counted once per homepage, including nested nodes. Only homepages were measured; product and article pages usually carry more markup. Our crawler identified itself as Mentioned-Research/1.0. Sites that refuse unknown bots at the firewall are missing from the sample, and those sites are probably stricter with AI crawlers too, so treat every blocking rate here as a floor rather than an estimate of the whole web.

Cite this

You are welcome to quote and chart these figures with a link back. Suggested citation: Mentioned, “Structured Data on the Top 1,000 Homepages: What AI Crawlers See,” October 28, 2026, https://mentioned.at/learn/homepage-structured-data-study. Data and text are published under CC BY 4.0.

Questions

Run a free check