How ready is your site for AI search?
We sample up to 10 of your pages and score four things models actually depend on: can they crawl you, can they parse you, can they quote you, and does your technical setup get in the way. No sign-up.
What the four sub-scores measure
The composite is their plain mean. Each one is checkable by hand, which is the point — you should be able to disagree with a number and go look.
Crawlability
Whether your robots.txt lets the 13 AI crawlers we track reach your site root. A single Disallow line here caps everything else at zero value.
Content quality
Per page: one h1, real subheadings, a short opening paragraph a model can quote, readable prose, and enough depth to be worth citing.
Structured data
Whether each sampled page ships valid JSON-LD, and whether any of it is a type that earns rich results — Article, FAQPage, Product, BlogPosting, NewsArticle.
Technical SEO
robots.txt and sitemap.xml parsed for real errors, plus crawl-budget waste: orphan pages, sitemap URLs that 404, and URLs listed in the sitemap that robots.txt forbids.
A 10-page sample is not a site audit
We say this on the report too. The tool reads your homepage and the first same-origin URLs your sitemap lists, up to 10 pages, and every report names the exact count it managed to read. That is enough to catch the template-level problems that account for most AI-readiness failures — a missing h1 pattern, no JSON-LD anywhere, a robots.txt that blocks GPTBot. It is not enough to tell you that page 4,000 is fine.
When a crawl reads nothing at all, you get no score rather than a zero. An unreachable site is an unknown, and reporting an unknown as a failing grade would be the kind of thing this tool exists to catch other people doing.
Just want the crawler question answered?
If you only need to know whether your robots.txt is blocking ChatGPT, that check is instant and needs no crawl — a red/green matrix of all 13 AI crawlers.
Check my AI crawler accessAI readiness questions
How many pages does this actually check?
Up to 10, and the report states the exact number it managed to read. This is a sample, not a full-site audit. We take your homepage plus the first same-origin URLs in your sitemap.xml. On a 5,000-page site it is a spot check — a useful one, because the problems it finds are almost always site-wide template problems, but it cannot tell you about the 4,990 pages it never saw.
What is the score out of 100?
The plain mean of the four sub-scores: crawlability, content quality, structured data and technical SEO. No weighting, no secret formula. If a crawl reads nothing, you get no score at all rather than a zero — an unreachable site is an unknown, not a failure.
Is the report public?
Each run gets its own URL you can share. Reports from crawls that genuinely produced findings are indexable; reports from failed or empty crawls carry a noindex tag, because a page saying "we could not read this site" has no business in a search index.
Does a high score mean ChatGPT will cite me?
No. This measures whether the mechanics are in your favour: can the crawlers reach you, is your markup parseable, is your writing quotable. Whether a model then names you depends on your authority and how often you are referenced elsewhere — things no crawler can score in 15 seconds. Be suspicious of any tool that claims otherwise.
What do you do with the domain I enter?
We request a handful of public URLs on it with a GrowzoBot user agent, exactly as any crawler would, and store the resulting report. We only fetch public internet hosts over ports 80 and 443, and we re-check every redirect hop — so the tool cannot be pointed at a private network or a cloud metadata endpoint.
Why do you ask for an email for the full findings?
The score, all four sub-scores and the most serious findings are free and need no email — that part is public and indexable. The full list, which on a typical site runs to a few dozen items, is behind one email field. If you would rather not, the free half is still the honest half: nothing is hidden that changes the verdict.
A score is a snapshot. Visibility is a habit.
Growzo watches your crawler access, publishes the content models want to cite, and tracks whether ChatGPT, Claude and Perplexity actually name you.
Open-source · Self-host · BYOK · ₹10k/mo hard cap