Canonical URL

Every page you want in search results should declare one authoritative web address for itself, so search engines know which copy of the page to index when the same content is available at more than one address. Silktide checks that each page declares exactly one valid address of this kind.

The canonical URL is set with a <link rel="canonical"> tag in the page's HTML header:

<!-- A valid, self-referencing canonical -->
<link rel="canonical" href="https://www.example.com/about">

A valid canonical is a single, absolute URL on the same site as the page. Common problems include a missing tag, more than one tag with conflicting targets, a relative or malformed URL, and a canonical that points at a different domain.

Why this matters

Websites often serve the same content at multiple URLs - with and without www, with tracking parameters, or over both http and https. Without a canonical, search engines must guess which version to rank, which can split ranking credit between duplicates or index the wrong copy.

A broken canonical can be worse than none. A canonical that points at another domain tells search engines not to index your copy of the page at all, and conflicting canonicals are simply ignored, silently wasting the tag.

How to fix it

1. Add a <link rel="canonical"> tag to the page header. For most pages it should point at the page's own preferred URL. Most content management systems and SEO plugins can generate this automatically.
2. Use a complete, absolute URL including the scheme and domain (https://www.example.com/about, not /about), with no surrounding whitespace.
3. Declare exactly one canonical per page. If your templates and an SEO plugin both add one, remove the duplicate.
4. If the canonical deliberately points at another domain (for example, syndicated content whose original lives elsewhere), that is legitimate - approve the finding.

How Silktide tests this

1. Test only pages that are eligible for search indexing (pages that are noindex, behind a login, or returning errors are skipped).
2. Collect every <link rel="canonical"> tag on the page. This is a structural check of the page itself; the canonical's target is not fetched.
3. Report at most one finding per page - the most significant problem, in this order:
   • No canonical tag at all.
   • More than one canonical with different targets.
   • A canonical that is not a valid absolute URL.
   • A canonical that points at a different domain than the page (www and non-www count as the same site).

Troubleshooting

My canonical is a relative URL and search engines accept it

Search engines usually resolve relative canonicals, but they are fragile: a base URL change or a page served at an unexpected address silently changes what the canonical points at. Silktide requires an absolute URL because it is unambiguous.

The canonical points at another domain on purpose

Cross-domain canonicals are valid for syndicated content. Silktide flags them because they are more often a copy-paste mistake than a deliberate choice; approve the finding if yours is intended.

Learn more

• Consolidate duplicate URLs (Google Search Central)
• The canonical link type (MDN)
• Canonical vs noindex