Titles and descriptions measured in pixels rather than characters, because that is how Google truncates them. Plus canonical, indexability, redirect chains, heading order and whether the URL is genuinely in your sitemap.
Most on-page problems are invisible when you look at a page in a browser. A noindex delivered in an HTTP header renders nothing. A canonical pointing at another URL looks identical to one pointing at itself. A title that fits in characters can still be cut in the results.
An X-Robots-Tag noindex is sent as an HTTP header. View-source shows nothing. Pages disappear from the index and nobody can find the cause.
Title guidance is usually given in characters. Google truncates on pixel width, so capitals and wide letters cut earlier than a character count suggests.
A page can be perfect and still not rank if it is absent from your sitemap and has no internal links pointing at it.
Character count and measured pixel width. Flagged above 580px for titles and 990px for descriptions, where truncation begins.
Robots meta and the X-Robots-Tag header. The header version is the one that catches people out.
Whether the page points at itself or elsewhere, with the destination shown when it differs.
Follows up to five hops and lists each one. Every hop costs crawl budget.
H1 count and whether levels skip. One H1, no jumps from H2 to H4.
Finds your sitemap via robots.txt or the usual paths, follows index children, and checks this URL is listed.
Counts JSON-LD blocks and validates that each one parses. Invalid JSON is ignored entirely by search engines.
How many images carry alt attributes and how many do not.
og:image, twitter:card, viewport and the lang attribute.
Enter a live page. We fetch it as an ordinary browser would, follow any redirects, then check the twelve items and show the fix for anything that fails.
Also fetches robots.txt and your sitemap, so allow a few seconds.
The check found what it expected. Nothing to do. A page can pass every check and still not rank — these are prerequisites, not a ranking recipe.
Present but outside the range that works reliably. A title at 620px will still rank; it will just be cut in the results. Worth fixing, not urgent.
Something is actively preventing the page from working. A noindex, a missing H1, invalid JSON-LD. Fix these before anything else on the page.
The score is a weighted summary and exists only to give you a sense of scale. The individual checks are what you act on.
Every check here is mechanical. It can tell you a page is technically able to rank. It cannot tell you the page deserves to, and no automated tool can.
Core Web Vitals need field data from real visitors. Use PageSpeed Insights for that. Structure and speed are separate problems with separate fixes.
If your content renders entirely in JavaScript, this sees what a crawler sees before execution — which is usually exactly the thing worth knowing.
Close to how Google renders but not identical. Treat 580px and 990px as guidance rather than a hard cut-off.
The score reflects our judgement of what matters. Anyone reasonable could weight it differently and get a different number from the same page.
Because Google truncates on width. A 60-character title in capitals is wider than a 70-character title in lower case, so a character count tells you very little about whether it will be cut.
Indexing is not instant, and removal is not either. A page can remain in the index for weeks after a noindex appears. If this reports one, the page is on its way out.
For one page, rarely. Across thousands it wastes crawl budget on hops rather than pages. The fix is cheap: link to the final URL.
It usually does not for a well-linked page. It matters most for pages that are new, deep in the structure, or have few internal links — exactly the pages most likely to be missed.
No. Fix the failures, consider the warnings, and spend the rest of your time on the content and the links. Structure gets you eligible, not ranked.
Every finding here has a fix attached, and most of them are quick. The harder question is which of your pages have these problems, and which of those are worth fixing first.