SEO Audit Tool

An on-page audit inspects the HTML that was returned: title, description, headings, links, images, and similar elements. PageSpeed adds lab metrics for that URL when the request succeeds. The report is for the single URL you submit.

The report groups errors, warnings, and passed checks from the HTML that came back. Open a failed row to see what the fetch found. PageSpeed details appear when that request returns. If the host blocks the fetch, the report can only describe the response it got.

Fetched HTML is one of the values SEO Audit Tool puts on screen. The audit reads the document the server got back. Title, description, headings, links, and images are judged from that HTML. Client-rendered text that was not in the response will not be seen.

Read Errors on its own before you mix it with the other rows. Failures the checks treat as problems. They are listed in the report. Fix the HTML, then fetch again. The report does not edit the site.

Warnings answers a narrower question than the headline number. Issues that are weaker than errors. They are grouped separately. A warning is not a ranking promise.

Treat Passed checks as a label with a specific job. Checks that matched the HTML. They show what was present. A pass is about that response, not about the whole site.

The PageSpeed line is worth a full stop. A lab request for the same URL. Details appear when the request returns. If it fails, the HTML checks can still be there. This is separate from the sample Page Speed Test UI.

Use it when you want the server to retrieve one URL. Do not use it as a whole-site crawler, a rank tracker, or a substitute for Search Console. The fetch sees the HTML the server received. Content that appears only after a browser runs scripts may be missing.

If you remember one sequence from SEO Audit Tool, remember the fields in the order they change a decision. Fetched HTML matters because The audit reads the document the server got back. In practice, Title, description, headings, links, and images are judged from that HTML. The mistake to avoid is this: Client-rendered text that was not in the response will not be seen. Errors matters because Failures the checks treat as problems. In practice, They are listed in the report. The mistake to avoid is this: Fix the HTML, then fetch again. The report does not edit the site. Warnings matters because Issues that are weaker than errors. In practice, They are grouped separately. The mistake to avoid is this: A warning is not a ranking promise. Passed checks matters because Checks that matched the HTML. In practice, They show what was present. The mistake to avoid is this: A pass is about that response, not about the whole site. PageSpeed matters because A lab request for the same URL. In practice, Details appear when the request returns. The mistake to avoid is this: If it fails, the HTML checks can still be there. This is separate from the sample Page Speed Test UI. After that, the checks are simple. The URL is the page you mean, not the whole host. You waited for the fetch. You opened failed rows before changing templates. You did not treat a pass as a ranking. A blocked response was read as the response, not as the real page.

The worked example for SEO Audit Tool, read as one scene, is this. Submit https://example.com. The server requests that URL, analyzes the HTML it receives, and requests PageSpeed. You see errors, warnings, and passes for that response, not for every URL on example.com. If the host returns a block page, the audit describes that block page. It cannot invent the HTML it did not receive.

Keep SEO Audit Tool as this step only. When the job moves on, the Page Speed Test UI is the next page: Page Speed Test UI shows a performance score and lab-style metrics for a URL you type. The Page Size Checker covers a different piece of the same work: Page Size Checker asks for a URL and then shows a total kilobyte figure plus a split across HTML, images, scripts, and CSS.