The canonical tag is a <link rel="canonical" href="..."> element in your <head> that tells search engines: if this page exists at multiple URLs, treat that other one as the master.
Why it exists
The same content can be reachable at many URLs:
example.com/page,example.com/page/,www.example.com/page(slash variants)example.com/page?utm_source=newsletter(tracking params)example.com/products/red-shoeandexample.com/products/red-shoe?ref=top(parameters)- HTTP and HTTPS variants during migration
- Paginated, sorted, or filtered category pages
Without a canonical, Google has to guess which URL to index, dilutes ranking signals across the variants, and may pick the version you don't want.
When to use it
- Self-canonical: every page should canonical to itself. Belt-and-suspenders against parameter-based duplication.
- Cross-domain canonical: syndicated content should canonical back to the original publisher.
- Parameterised pages: tracking-tagged URLs canonical to the clean URL.
Common mistakes
- Canonicalling to noindex pages — Google ignores the canonical.
- Canonical chains — A canonicals to B which canonicals to C. Resolve to the final URL directly.
- Conflicting signals — page is
noindexbut canonical points elsewhere. Google ignores the noindex and may rank an unintended URL. - Missing canonical on filter/sort pages — common cause of cannibalization in ecommerce.
Canonical is a hint, not a directive. Google may override if other signals conflict.