Nine checks
Organization entity statement
- What
- Whether the homepage source contains a machine-readable "who we are" statement (Organization structured data).
- How
- View page source and search for application/ld+json; or run the Website AI Readability Check.
- Pass
- Company name, official URL, logo and sameAs links are present and match what users actually see.
Certifications as text
- What
- Whether certification and test-report numbers appear in the page text layer, not only inside images.
- How
- Press Ctrl+F on the page and search the number; if not found, machines likely cannot read it either.
- Pass
- Numbers appear as text with their source noted (standard number, lab, or report page).
Site vs. platform consistency
- What
- Whether prices, specifications and the company name match across your site and platform profiles.
- How
- Build a field-by-field consistency table between the official site and each platform.
- Pass
- Key fields match; genuine mismatches are explained with a stated source of truth.
- Tool
- Manual comparison — not covered by the free tools
Every paragraph stands alone
- What
- Whether a paragraph still makes sense — subject and all — when read out of context.
- How
- Sample key paragraphs yourself; or inspect chunking with the Page Chunk & FAQ Structure Check.
- Pass
- A lifted paragraph still identifies who is speaking (self-contained paragraphs).
FAQ self-containment
- What
- Whether each Q&A still identifies who is answering when lifted out alone.
- How
- Read each answer: where "we" first appears, is the entity named?
- Pass
- Every answer names the entity at least once; question and answer are complete pairs.
- Tool
- Website AI Readability Check (FAQ structure item)
Title and description
- What
- Whether the homepage and key pages each have a unique title and description.
- How
- The tab shows the title; the description sits in a meta tag in the source, or use the check tool.
- Pass
- No duplicates, no blanks, no keyword stuffing; the description matches the page.
Language tags
- What
- Whether the declared language matches the actual content, and multilingual sites carry proper tags.
- How
- Check the lang attribute and hreflang in the source; or use the check tool.
- Pass
- Declarations match reality; one page does not carry contradictory declarations.
robots.txt and sitemap.xml
- What
- Whether both files exist, are reachable, and do not block the public pages machines should see.
- How
- Open your-domain/robots.txt and your-domain/sitemap.xml directly in a browser.
- Pass
- Public pages are listed in the sitemap; robots.txt does not over-block; internal paths are properly excluded.
llms.txt
- What
- Whether llms.txt is provided and states the brand, business and canonical sources.
- How
- Open your-domain/llms.txt directly in a browser.
- Pass
- The file exists and matches the site. Note: llms.txt is a community proposal with no verified evidence that mainstream AI products read it; cheap to add, but no magic bullet.
After ticking the boxes
Fill gaps by priority: first the organization statement and certifications-as-text — they decide whether machines can identify you — then the rest.
Re-test under identical conditions and keep a before/after comparison. Complete signals do not mean AI will recommend you; that requires real model answers, which are a different class of evidence.
Further reading: the same material is also published in our Zhihu knowledge base (self-published, not independent evidence). Open the knowledge base
Free tools, no sign-in needed