Is your site ready for an AI chatbot?
A chatbot trained on your website is only as good as what a crawler can read from it. We sample your pages exactly the way our indexer does and report what it finds — sitemap coverage, pages that render only in the browser, and how much readable text each page really has.
Free, no signup, no limits worth mentioning.
It runs the real indexer, not an approximation
The scan calls the same functions the production crawl calls to fetch a page, extract its readable content, and count its words. That is the point of it: the verdict a stranger gets here is the verdict they would get after signing up, rather than a demo tuned to look encouraging.
The three checks
Publishes a sitemap — whether one is declared in robots.txt or findable at the usual paths, and how many URLs it leads to. Without one, pages have to be listed by hand.
Pages render on the server — whether the initial HTML already contains the content, or arrives as an empty shell that JavaScript fills in later. Shell pages can still be indexed, but need a real browser to do it, which is slower and misses more.
Pages carry enough text — how much readable content is left once navigation, headers and footers are stripped out. A page of four sentences is not something a chatbot can answer from.
How the sample is chosen
Up to eight pages, spread evenly across your largest sitemap rather than taken from the top of it. Sitemaps are usually ordered, so the first eight entries are typically the home page and its immediate children — the least representative pages on any site.
URLs that are obviously not content, such as cart and login pages, are filtered out before sampling, so the result reflects the pages that would really be indexed.
The credit estimate
The median word count of the sampled pages is extrapolated across every URL in your sitemap, using the same words-per-credit figure that billing uses. It is shown as an estimate because that is exactly what it is — a sample of eight pages standing in for the whole site.
Common questions
What does chatbot-ready actually mean?
That a crawler can find your pages and read real text from them. Those two things decide almost everything about how good a chatbot trained on your site will be, and both are fixable once you know which one is failing.
How many pages does the scan look at?
Up to eight, spread evenly across your largest sitemap so the sample is representative rather than eight versions of your home page. The whole scan takes well under a minute.
My site is a single-page app. Will it fail?
It will be flagged as rendering in the browser rather than on the server, which is a real finding but not a blocker. Those pages need a full browser render to index, which is slower and occasionally misses content, so server-rendered HTML remains the better outcome.
Does running this create an account or store my site?
No. The scan runs on request, the result goes straight back to your browser, and nothing is kept afterwards. No signup, no email address.
Do I need a sitemap for a chatbot to work?
It is by far the easiest way, because it hands over a complete list of your pages at once. Without one, content has to be added by pasting text or uploading documents instead.
Now let it answer your visitors' questions
TeklTalk reads the same pages you just checked and answers questions about them, with a link back to the page each answer came from.
Free to start. No credit card required.