What we observe
For each page we read, we compare the text in the HTML the server returned with the text a browser shows once the page has finished loading. A page depends on JavaScript for its main wording when 40% or less of the finished page’s words are in the server’s HTML and the finished page has at least 100 words. We also note links and head tags that appear only after scripts run. Anyone can repeat this by loading the page with JavaScript turned off.
Why it is here: the evidence
Measured An independent study observed it. No platform’s own document says so.
Source: Vercel and MERJ’s study of AI crawler fetches: the crawlers fetch JavaScript files but do not run them. Source grade A: read at the source: the document, the code or the study itself.
Last re-checked ; due again within 30 days.
What we cannot see
Which crawler runs JavaScript on which fetch. Google renders JavaScript; the study behind this check found that the AI crawlers it measured fetch script files without running them. We report the mechanism, never who is affected.
How to fix it
Serve the main wording in the HTML itself, through server-side rendering, static generation or prerendering, so the page reads in full with scripts turned off.
- Layer
- Read: Can an AI crawler fetch your pages and understand what is on them?
- Level
- One of the checks of Level 1: Can an automated visitor that says who it is get your pages and read them?
- Where its finding lives
- Search & findability
- How common it is
- Not published yet. We publish a share only once at least 30 sites stand behind it.
- Check id
chk-render
Cite as: Invara, “How AI sees your site”, method version 1, check chk-render. https://clearsitedigital.com/how-ai-sees-sites/chk-render