Enter one web address. Nine AI systems fetch that page the way they really do — no browser, no JavaScript — and you see exactly what each one receives. Under a minute, and it costs nothing.
No account. No sales call. One test per email address.
Enter one web address to test
By running the test you agree to our terms of use.
Not from what your website shows a person. From what its crawler was handed — once, with no browser, and nothing waiting for the page to finish building.
Those are usually two different pages. This shows you both, side by side, in under a minute.
Every one of these was received in full, or near enough, by all nine AI systems — a strong score on the question most tools stop at. Then we measured whether any of them could use what arrived. Not one cleared half.
Nothing told a machine how to book or buy.
Every word arrived. None of it said who the firm is for.
Structured data present — none of it identified the business.
No structured data on the page at all.
No machine-readable address, and no opening hours.
A machine could not tell what the practice does.
Nothing said where the firm is, or how to start.
No address, no hours, and nothing to book.
Swipe for more →
Arriving is not the same as being usable. A page can be delivered perfectly and still give an assistant nothing it can quote. That gap is where the enquiry goes to someone else, and it is the thing almost nobody is measuring.
Most ways of checking a page use a browser. It runs your JavaScript, waits for everything to assemble, then reads the finished result. That is a fair picture of what a person sees.
It is not what an AI crawler gets. GPTBot asks for the page and takes what comes back. It runs nothing, waits for nothing, and does not come back for the rest.
Your report shows both numbers, so you can see the difference on your own page rather than take our word for it.
Two scores out of 100, and the reason behind each.
A link to your full report, and a PDF you can keep. No dashboard to learn — a short report you can hand to whoever owns the website.
Each one identifies itself differently and behaves differently, so each is sent the request its own crawler sends — one after another, at the same moment, rather than treated as interchangeable.
It does not tell you whether AI is recommending your business. It asks no AI platform a question.
It reads one page — not your whole site, not your Google Business Profile, not anywhere else you appear.
What it does tell you is whether a machine can read you at all. That is the floor. Everything else depends on it.
Yes. One test per email address, no call, nothing installed, nothing to cancel.
No. SEO asks whether Google will rank a page for a search. This asks whether an AI system can retrieve and use the page at all — different crawlers, a different request, and a different bar. A page can rank well and still hand an assistant nothing it can quote.
Your first score will probably be high, and that is worth confirming rather than assuming. The second is the one that surprises people. Server-rendering gets your words to the crawler; it does nothing about whether your address, hours, services and booking path are published as data a machine can read. Most sites that deliver perfectly still fail that.
Partly, and you should. You can fetch your own page without a browser and read what comes back. What takes longer is sending as nine different operators, checking each against your robots.txt, and judging the result against what an assistant needs before it will quote you. That is the part this does for you.
We send your report. After that, only if you tick the box asking us to — and every email has an unsubscribe link.
Most are. The report tells you what is wrong and what it would take to fix it, whether you fix it yourself or not. There is no call attached — the report arrives either way.
Either a machine can already read your business, or you find out exactly what it cannot.
Already know you need this? See the three ways to fix it →
We tested nine AI operators against your page. Enter your email to view the result.