Canonical analysis

The Canonicals tab in the analysis panel inspects every crawled page for its rel=canonical tag, surfaces pages where the tag is missing or points somewhere unexpected, and provides a statistical overview of how the site uses canonicalisation.

Stats

The top of the Canonicals tab displays a row of summary statistics drawn from the full crawl:

Stat	What it counts
Total pages	All HTML pages in the crawl that returned a 2xx response.
Pages with canonical	Pages that contain a `rel=canonical` tag (in the `<head>` or via the `Link` HTTP header).
Self-referencing canonicals	Pages whose canonical URL resolves to the page itself. This is the recommended configuration.
Cross-domain canonicals	Pages whose canonical points to a different registered domain. Common in syndication, but worth reviewing.
Missing canonicals	Pages that have no `rel=canonical` tag at all.

On a well-configured site, the self-referencing count should be close to the total page count. A large gap between the two is a signal worth investigating — it means many pages either lack the tag entirely or canonicalise to a different URL.

Issues

Below the stats, the tab lists every page where a canonical-related issue was found. Each row shows a severity-coloured left border, an issue type badge, the page URL, and a short message explaining the finding.

The issue types are:

Issue type	Severity	Rule	When it fires
Missing canonical	Warning	`canonical_missing`	The page has no `rel=canonical` tag. Without one, search engines decide on their own which URL to index — and they may choose a version with query parameters, a trailing slash variant, or a duplicate you did not intend to surface.
Canonicalised elsewhere	Notice	`canonical_non_self`	The page has a canonical tag, but it points to a different URL on the same domain. This is intentional when consolidating duplicates, but it also appears when a CMS sets the wrong canonical during a migration or template change. Review the target URL to confirm the signal is deliberate.
Cross-domain canonical	Informational	—	The canonical points to a URL on a different registered domain. This is expected for syndicated or republished content, but it effectively tells search engines to drop the page from your index in favour of the remote version. Flagged for visibility rather than as a problem.

Issues are sorted by severity: warnings first, then notices, then informational. Within each severity level they are ordered by URL. Click any row to open the page in the detail inspector.

How canonicals work

The rel=canonical tag tells search engines which URL is the preferred version of a page. When multiple URLs serve identical or near-identical content — because of query parameters, session IDs, tracking codes, or print-friendly variants — the canonical tag points them all at a single authoritative URL. Search engines consolidate their signals (links, ranking data) onto that URL instead of splitting them across duplicates.

A self-referencing canonical is best practice: each page declares itself as the canonical version. This removes ambiguity and prevents search engines from choosing a different URL on your behalf. Even pages with no duplicates benefit, because query parameters appended by analytics tools or ad platforms can create unintended variants.

When the tag is missing, the choice falls to the search engine. It will pick the URL it considers most representative based on internal signals — which may not match what you intended. A missing canonical is not an error in the way a broken link is, but it is a gap in the instructions you are giving to crawlers.

A canonical that points to a different page on the same site is a deliberate statement: this page is a duplicate, and the target is the one to index. That is the correct use when consolidating faceted navigation pages, paginated series, or A/B test variants. It becomes a problem when it happens by accident — a template that hardcodes a canonical, a migration that maps URLs incorrectly, or a staging environment whose canonicals leak into production.