See a page the way an AI crawler gets it: no JavaScript, just the HTML. Check its JSON-LD, the robots tags that keep it out of answers, and how much of the page is actually there.
application/ld+json block: whether it parses (a block with a trailing comma is ignored entirely), which schema.org types it declares, including inside @graph, and which of the properties each type usually carries are missing.noindex, nosnippet and max-snippet:0, in a robots meta tag or the X-Robots-Tag header.<div id="root"> until JavaScript fills it reads as nearly empty to a crawler that doesn’t render.The “usually carries” lists come from schema.org and Google’s structured data documentation. They’re a completeness check, not a ranking formula. Nobody outside the model companies knows exactly how markup is weighed in an answer.
Organization block on the homepage: name, URL, logo, and sameAs links to your LinkedIn, G2 and Crunchbase profiles. This is how a parser ties every page back to one company.SoftwareApplication or Product on the pages that say what you sell, with an offers block that matches your pricing page.Article with an author and real datePublished and dateModified values on guides, so a reader can tell how current the page is.Then check that crawlers can get in at all with the AI crawler checker.
There’s no published evidence that it does on its own. It does make the facts on a page unambiguous to any parser that reads it, and Google uses it for rich results. Our guide on schema markup and AI citations goes through what is and isn’t known.
Because most AI crawlers do. Vercel’s December 2024 study of crawler traffic found that the major AI crawlers fetch JavaScript files without running them. Text and JSON-LD that only appear after a client-side render are invisible to them.
Organization on the homepage, with your name, logo, URL and sameAs links to your profiles. SoftwareApplication or Product where you describe what you sell, with a real offers block. Article on guides, with an author and dates. FAQPage only where the questions and answers are visible on the page.
Google says AI Overviews and AI Mode follow the same preview controls as Search. A nosnippet robots tag, or max-snippet:0, stops Google showing text from the page, so it can’t be quoted there either. The checker reads both the robots meta tag and the X-Robots-Tag header.
No. Google’s test tells you whether a page is eligible for Google’s rich results, and it renders JavaScript. This one reads the raw HTML a non-rendering crawler gets and checks completeness against the properties each type usually carries. Use both.
Clean markup makes your pages easier to read. It doesn’t tell you whether ChatGPT, Claude or Gemini recommend you when a buyer asks, or what they say about your competitors instead. Mentioned does: it asks the questions your buyers ask, the same way every week, and ranks what to fix first.