What it reports, and what each finding means
- Title width in pixels against the 600 Google shows. Over that, the end is cut.
- Description width against 920. Same rule.
- Canonical — present, and pointing where you expect. A canonical to another page means this one is asking not to be indexed in its own right.
- Robots —
noindexhere is the single most consequential tag on the page, and the most commonly left behind after a staging deploy. - hreflang count — zero on a multilingual site means the annotation is missing.
- Images without alt — a cheap accessibility finding that is also a signal.
The check worth running first
noindex. It is a single word in a meta tag that removes a page from search
entirely, it produces no visible symptom on the page, and it reaches production
regularly — carried over from a staging environment or left by a plugin.
If traffic to a page vanished and nothing else changed, check this before anything else.
It sees what a crawler sees
The page is fetched anonymously, without running JavaScript. If your meta tags are set client-side, they will not appear here — and for most crawlers they do not appear either. That absence is a finding, not a limitation of the tool.
Questions
Why does this need an account?+
Because it makes a request from our server on your behalf, which costs real resources and can be abused. Every other tool on this site runs entirely in your browser and needs nothing.
Can it read a page that requires a login?+
No. It fetches the page as an anonymous visitor, so it sees what a search engine sees — which is usually the point. A page behind a login is a page search engines cannot index either.
Why does the title width differ from the character count?+
Because Google truncates by rendered width. The pixel figure is what determines whether your title is cut; the character count is a proxy that fails on titles made of wide or narrow letters.
It says the address is not publicly reachable.+
The tool refuses private and internal addresses — localhost, 10.x, 192.168.x, cloud metadata endpoints. A server that fetches any URL it is given can be pointed at internal services, so the check is deliberate and applies to every redirect hop too.