Free tool · No account
SEO & GEO audit
Paste a page's address and see how it does on the 25 checks Quaryan runs for its customers: 16 for Google and 9 for AI answers like ChatGPT. Each one is measured on the page itself, so the answer is the same every time. Free, no account.
How it works
How the SEO & GEO audit works
01
It reads the page like a crawler
Quaryan fetches the page the way a search engine does, following redirects and without running JavaScript, then the site's robots.txt, its sitemap and its llms.txt.
02
It makes the checks Quaryan's customers get
The same 25 checks a Quaryan account runs, each measured rather than guessed: a title is 62 characters or it is not, and robots.txt blocks GPTBot or it does not. Ask twice and you get the same answer.
03
It shows what it found
Each check is Good, Worth fixing, a Problem, or Not checked when the site did not answer, with what is on the page: the title, the headings, the crawlers that are blocked.
04
The fixes are in a free account
Which of them matter for your business, in what order, and the change to make for each. A free Quaryan account writes them, checks the page again once you have made them, and runs the audit every week.
The checks
The 25 checks, and why each one matters
Sixteen for search engines and nine for AI answers. The first two count for both: a page that may not be indexed, or has almost nothing on it, is lost to either.
| Check | Why it matters |
|---|---|
| For Google | |
| Indexing Google and AI | A robots meta tag, or an X-Robots-Tag header, can tell search engines to leave a page out altogether. |
| How much is on the page Google and AI | There has to be enough here to rank, and enough for an assistant to quote. |
| Served securely | A page served over plain HTTP is marked as not secure in the browser, and ranked below one that is. |
| Title | The line a search result is headed with. Google shows about sixty characters of it. |
| Meta description | The two lines under a search result, and the only sales copy you control there. Seventy to a hundred and sixty characters get shown. |
| Main heading | One top-level heading, saying what the page is about. |
| Subheadings | Subheadings break a long page into the questions it answers, for a reader and for a crawler. A short page can do without. |
| Canonical URL | When a page can be reached at several addresses, a canonical URL says which one counts, so its ranking is not split between them. |
| Mobile viewport | Without a viewport tag, a page is laid out at desktop width on a phone, where most of its visitors are. |
| Declared language | The lang attribute tells a search engine which market the page is written for. |
| Image alt text | Alt text is what an image says to a crawler and to a screen reader. |
| Links to your own pages | Links to the rest of the site are how a crawler finds it, and how ranking moves between pages. |
| robots.txt | Where a site says what may be crawled. A blanket Disallow turns everyone away, including the search engines you want. |
| Sitemap | A sitemap lists the pages worth indexing, so a crawler does not have to find them all by following links. |
| Link previews | Open Graph tags decide what a link to the page looks like when someone shares it: its title, its description and its image. |
| Structured data | Structured data restates in machine-readable form what the page already says in prose. |
| For AI answers | |
| AI crawler access | Answer engines send crawlers of their own, and a robots.txt that turns them away opts the site out of being an answer. |
| Readable without JavaScript | Most answer engines do not run JavaScript when they fetch a page, so what counts is what the HTML itself carries. |
| Who and what you are | Structured data naming the organization, the product or the service gives a model facts to state rather than prose to paraphrase. |
| Questions and answers | FAQ structured data pairs a question with a stated answer, which is the shape an answer engine is looking for. |
| Headings that answer questions | Headings phrased the way people actually ask are what a model matches a question against. |
| Figures worth quoting | A model asked what something costs needs a figure on the page to quote. |
| When it was last updated | A date, visible or in the markup, lets a model say how current the page is, and whether to trust it over something newer. |
| Who is behind it | An about page, a named author or a publisher in the structured data: what a model weighs when it decides whom to cite. |
| llms.txt | An /llms.txt describes a site to language models in plain text, the way robots.txt describes it to crawlers. |
Audit this page
Free · No account · A few seconds
FAQ
Questions about SEO and GEO audits
What does this SEO audit check?
Twenty-five things a search engine or an AI assistant reads on a page. For Google: whether the page may be indexed, how much is on it, HTTPS, the title, the meta description, the headings, the canonical URL, the mobile viewport, the language, image alt text, internal links, robots.txt, the sitemap, link previews and structured data. For AI answers: whether AI crawlers are let in, whether the page reads without JavaScript, structured data saying who you are, questions and answers, headings phrased as questions, figures worth quoting, a date, who is behind the site, and an llms.txt.
What is GEO, and how is it different from SEO?
SEO, search engine optimization, is the work of ranking in Google's list of links. GEO, generative engine optimization, is the work of being named in the answers AI assistants such as ChatGPT write. They share foundations, a page crawlers can read that says plainly what it offers, but an assistant also has to be let in by robots.txt, to read the page without JavaScript, and to find facts it can quote.
Is the SEO audit free?
Yes. It is free, needs no account and takes a few seconds. Each visitor can audit thirty pages an hour, and an answer is kept for ten minutes so a page is not read twice in a row.
How is this different from the audit in a Quaryan account?
The checks are the same. An account adds the other half: which failures matter for your business and in what order, the change to make for each (the tag, the markup, or a prompt for your developer or coding agent), a recheck once you have made it, and the audit again every week. All of that is in the free plan.
Why does a check say “Not checked”?
Because we could not find out. A robots.txt or a sitemap that did not answer is not the same as one that is missing, so the checks that need it are left unmade rather than reported as problems. Run the audit again in a minute.
My title looks fine. Why is it “Worth fixing”?
The audit counts characters: a title of 25 to 60 characters and a meta description of 70 to 160 fit most search results. Google cuts them by width rather than by count, so a title of wide letters can be cut sooner, and one of narrow letters later. Measure your title and description in pixels
If every check passes, will my page rank?
Not by itself. These checks are the foundations: they make sure nothing on the page stops a search engine or an assistant from reading it and understanding it. Whether it ranks, or is named in an answer, also depends on what it says, how well it answers the question, and who links to it.
Can I audit any page, or only my homepage?
Any public page. Paste its full address, like yoursite.com/pricing; a bare domain audits the homepage. The robots.txt, the sitemap and the llms.txt belong to the site, so they are the same for every page on it.
Next
The fixes, in the order worth doing them
A free Quaryan account turns this audit into a plan: what to change first on your site, the change itself, and the audit again every week.