Website SEO Analyzer

The scheme, hostname and port define the allowed origin. Use the site’s final address after redirects so the starting point matches the version you intend to inspect. A different subdomain or a change from HTTP to HTTPS can change that scope.

Page URLs cannot contain query strings, login or checkout paths, or downloads. Choose appropriate public page URLs for the task. Do not remove a query string merely to make a URL fit if doing so would identify a different page or change its meaning.

These restrictions keep the audit bounded. They also mean the report cannot represent every possible page variant, private application screen or resource on a site. Keep that boundary in mind when choosing a sample and describing its findings to someone else.

The visible form currently accepts a page limit from 1 to 10. The FAQ describes an administrator-selected ceiling of 20 outcomes. The current form’s limit applies to your run; the higher ceiling is not a promise of current availability.

Page outcomes, successful HTML inspections and HTTP requests are different counts. A failed page can occupy an outcome without producing metadata findings. Robots and sitemap requests also consume the shared bounded request budget.

Time, bytes, redirects and sitemap limits can stop work before the requested sample is fully inspected. A smaller completed sample needs a coverage explanation, not an assumption that every remaining URL is healthy.

Open a page’s evidence to review its returned title, description, headings, canonical declarations, robots instructions and relevant response details. Redirect evidence helps explain whether the requested page and final location differ.

The checks concern the returned source. A missing field means it was not observed within that analysis; it does not prove how every visitor or search engine saw the page. A failed response has a different evidential boundary from an inspected HTML page with missing metadata.

Use the source to identify the component that should change. A repeated problem across a template may have one shared cause, while two unrelated pages can need separate editorial fixes. A finding is most useful when it leads to a precise review rather than a general instruction to improve an SEO score.

The duplicate views compare nonempty metadata across distinct inspected final URLs, using case and whitespace normalization. Missing metadata remains a separate page finding.

Repeated wording can be intentional. Two pages might describe related versions of the same material, or a template might be repeating text that should be page-specific. Read both pages and their purpose before deciding which explanation applies.

A repeated title is not proof of duplicate body content. Similarly, changing a few words in a description does not establish that two pages now serve different needs. Use the grouping to find candidates for review, then make the editorial decision from the content and intended page relationship.

Review whether titles and descriptions clearly distinguish each page’s purpose. Length checks are review heuristics, not universal validity rules or guarantees of a particular search display.

A canonical declaration can point to the same page or another URL. The analyzer compares observed declarations and destinations within the inspected sample. Canonical destinations are not fetched automatically just because a declaration mentions them.

If a destination was inspected as part of the sample, the relationship has that page’s evidence available. If it was not inspected, retain the uncertainty. “Outside this sample” does not mean broken, missing or rejected by Google.

Use the Canonical Tag Checker when the declaration or destination needs focused investigation. Review the intended relationship before replacing a canonical with a self-reference. A legitimate alternate page can intentionally point elsewhere.

The report does not establish Google’s selected canonical. That question requires separate evidence about Google’s processing, rather than an inference from a small graph of declarations.

Robots instructions are checked before page retrieval. The audit crawler’s rules and Googlebot rules are interpreted separately. A page allowed for one crawler may have different instructions for another.

These are interpretations of technical declarations. They do not prove that a request came from Googlebot or that Google has indexed the page. A robots exclusion and a noindex directive also answer different questions and should not be treated as interchangeable labels.

Use the Indexability Checker to investigate related access and indexing-instruction evidence. Use Google’s own owner tools when you need information about its indexed version or selected canonical.

A sitemap is a source of declared URLs. Reading it does not establish that every entry was requested, returned usable HTML or received the same checks.

Sitemap mode has document, depth and entry caps as well as the shared request budget. Review source outcomes and discovered versus inspected URLs before interpreting the page sample. A source that could not be read should remain visible as a source limitation.

Use the XML Sitemap Checker when the XML, listed locations or sitemap structure needs a focused review. Sitemap submission itself is only a discovery hint to Google; it does not guarantee crawling or indexing.

The analyzer reads initial HTML and selected HTTP evidence. It does not execute JavaScript or load the page’s images, stylesheets and scripts as a browser would.

If a title, link or piece of text appears only after scripts run, the source-based sample may not contain it. Use the Rendered HTML Checker to compare original HTML with a rendered DOM snapshot you supply. Keep those evidence sources clearly identified.

Response timing describes the final HTTP request. It is not browser rendering time, a page-speed score or a Core Web Vitals measurement. A quick response cannot establish that an interactive page loads or responds quickly for real visitors.

Choose the view that fits the investigation: pages and findings, errors and warnings, duplicate titles, duplicate descriptions, or canonical relationships. Search narrows the displayed rows, and pagination controls the portion currently shown.

Display filters do not redefine the inspection’s coverage. A view with no matching rows can still belong to a report containing failed outcomes or findings elsewhere. Return to the overview and coverage notes before drawing a general conclusion.

Download CSV or JSON for the full available report, including source outcomes, input receipts, configured limits and incomplete states. Display filters do not remove evidence from downloads. A separate summary TXT gives you a concise review handoff. Failed runs can also be exported when source outcomes are available and downloads are enabled.

Some report sections can show unknown or unavailable coverage. That state is useful information: the report does not have the evidence needed to make the associated conclusion.

For example, HTTP resource references found in initial HTML are not a browser verdict about which mixed-content requests loaded or were blocked. Likewise, wording that resembles a not-found page is not itself Google’s soft-404 determination.

Do not turn an unavailable section into a zero-finding result. Keep its scope and limitations attached to the report. A separate targeted check may answer the remaining question more effectively than repeating an unchanged sample.

No. It inspects a bounded sample, currently up to 10 page outcomes. The administrator can configure a maximum of 20. Failed or excluded outcomes count toward the selected limit. Robots, request, time, byte and sitemap limits can reduce successful inspections.

No paid API key is required. Human verification and shared daily completed-check limits apply. The built-in verification code expires after ten minutes and works once.

No. It analyzes initial HTML and selected HTTP response evidence. Final-response timing and audit collection time are not browser rendering time or Core Web Vitals.

No. Googlebot robots rules, noindex directives and canonical declarations are technical evidence. They do not confirm Google indexing, Google’s selected canonical, rankings, traffic or domain authority.

No. It reports observations and comparisons. You decide which changes are justified and make them through your own publishing or development workflow.

The report retains failed, excluded and known unrequested outcomes separately. It does not invent page findings. CSV and JSON can retain available source evidence even when no page was inspected; display filters do not remove evidence from exports.

This tool does not create a saved report history. Keep the current tab open or download your report before resetting or leaving. No account, scheduled audit or comparison history is created by this analyzer.