How to use the SEO Checker
- Paste the full URL of the page you want to test, including
https://. - Run the check. We fetch the page once and run every test against that single response.
- Read your score, then open each of the four groups to see which checks passed, which need attention and which failed.
- Fix the failed checks first, then the warnings.
- Run the check again to confirm your changes are live.
Understanding your results
Your results are split into four groups.
AI readiness looks at signals that help machines, including AI systems, read and reuse your page. We check whether the page has JSON-LD structured data, whether your robots.txt blocks common AI crawlers (GPTBot, PerplexityBot, ClaudeBot, Google-Extended and CCBot), whether the site publishes an llms.txt file, and whether an Open Graph title and image are set.
Search basics covers the tags search engines read first: a title between 25 and 65 characters, a meta description between 70 and 165 characters, exactly one H1, a canonical tag and an html lang attribute.
Technical checks that the page loads over HTTPS, how quickly the server responds (under 800 ms is good, over 2.5 seconds is poor), whether the HTML is under 150 KB, whether a mobile viewport tag is set, whether a noindex directive is keeping the page out of search, and whether the response is compressed with gzip or Brotli.
Content checks for at least 300 words, alt text on images, at least three internal links and at least two H2 headings.
Every check returns pass, warn or fail. The score from 0 to 100 is the average of all checks, where a pass counts as 1, a warning as 0.5 and a failure as 0. It summarizes how many on-page basics are in place. It is not a prediction of rankings.
A few notes on specific results:
- The character ranges for titles and descriptions are practical guidelines, not rules. Google has no fixed length limit and shortens titles and snippets to fit the device width.
- A blocked AI crawler is flagged so you know about it, not because blocking is wrong. Allowing or blocking AI crawlers is a business decision. For example, Google-Extended controls whether content Google crawls can be used for Gemini models, and Google says it does not affect inclusion or ranking in Google Search.
- The one-H1 check reflects a common convention for a clear page structure rather than a Google requirement.
How this tool works
When you submit a URL, our crawler requests the page once and identifies itself as NewBacklinkBot. Some sites block unknown bots with a 403, 429 or 503 response. When that happens, we retry one time with a normal browser user agent, so you still get a result.
We also request the site's robots.txt and llms.txt files for the AI readiness checks, and we use the page's HTML and response headers to detect the platform it runs on, such as WordPress, Shopify or Wix.
Some pages send very little content in their HTML and build everything with JavaScript in the browser. Google renders JavaScript, but several AI crawlers do not, so what those crawlers see can be very different from what your visitors see. When we detect a page that is rendered mostly by JavaScript, we skip the content checks instead of reporting word counts and link counts of zero as failures, and we tell you why.
The SEO Checker tests one page at a time. It does not crawl your whole site, so run it separately on your homepage, key category or product pages and important articles. For a step-by-step review of an entire site, follow our SEO audit checklist.
Tips
- Start with failed checks in Search basics and Technical. A missing title, an accidental
noindexor a missing viewport tag can matter more than anything else on the list. - Review the exact tags that were found with the Meta Tag Checker, which shows character counts and a snippet preview.
- If the page has no structured data, add JSON-LD that matches the page type and test it with the Schema Markup Validator.
- If you want AI crawlers that do not run JavaScript to read a page, make sure its main content is present in the HTML your server sends.
- To publish an llms.txt file, create a draft with the llms.txt Generator & Checker and edit it before you upload it.
- Re-run the check after every change. Caching plugins and CDNs sometimes serve an older version of a page for a while.
- Test each page template your site uses, because a problem in a template repeats on every page built from it.
Frequently asked questions
Is the SEO Checker free?
Yes. The SEO Checker is free to use, and you can check as many individual pages as you need.
Does a score of 100 mean my page will rank first?
No. The score shows how many on-page basics are in place on one page. Rankings depend on many other things, including the quality of your content, how well it answers what people search for, and links from other sites.
Why did my page skip the content checks when it has plenty of text?
If your content is loaded by JavaScript after the page opens, the HTML we fetch may contain very little text. When we detect this, we skip the content checks rather than score them as failures. For important pages, consider server-side rendering or pre-rendering so the content is in the HTML.
Why does the tool check robots.txt for AI crawlers?
AI assistants and AI search features rely on crawlers to read web pages. If your robots.txt blocks those crawlers, the related services may not be able to read your site. We show what is blocked so that the choice is deliberate, not accidental.
Can the SEO Checker audit my whole website?
No. It checks one URL per run. Use it on your most important pages and templates, and use our technical SEO tools, such as the broken link checker and sitemap checker, for site-wide issues.
Why did the tool retry my page with a different user agent?
Some firewalls block bots they do not recognize. If your site answers our crawler with a 403, 429 or 503 status, we retry once as a normal browser. Your firewall may be blocking other legitimate crawlers too.