The essentials
- Definition: the canonical tag (rel="canonical") is an HTML element placed in the head that declares a page's reference URL when several addresses show the same content.
- Syntax: <link rel="canonical" href="https://www.example.com/page/">, with an absolute URL, one per page.
- Status: a strong hint, not a directive; Google can choose another canonical URL if it judges the signal inconsistent.
- Uses: URL parameters, e-commerce filters, HTTP/HTTPS or www/non-www versions, syndicated content, badly handled pagination.
Enhanced measurement canonical tag is a hint placed in a page's HTML head to designate, among several URLs with identical or very similar content, the one search engines should treat as the reference version. It answers the problem of duplicate content : without it, Google picks a version itself, splits the signals between the duplicates, and may index the wrong address.
Syntax and basic rules
The tag is written <link rel="canonical" href="https://www.exemple.fr/produit/chaise-bois/"> in the <head>section. Google also accepts the HTTP header Link: <URL>; rel="canonical", useful for PDF files. Five rules avoid most errors:
- Use an absolute URL (protocol and domain included), never a relative path.
- A one canonical tag per page; where there are duplicates, Google ignores them all.
- Point to a URL that returns 200, is indexable, and declares itself canonical (self-reference).
- Stay consistent with your other signals: sitemap, internal links, redirects and hreflang must all designate the same URL.
- Don't combine it with a noindex on the canonical page: the signal becomes contradictory.
The use cases
| Situation | Canonical to | Alternative |
|---|---|---|
| URLs with sorting, tracking or session parameters (?utm=, ?sort=price) | The clean URL without parameters | None; the canonical is the right answer |
| E-commerce filters generating thousands of combinations | The main category page | Robots.txt or noindex if the filtered pages have no value at all |
| http/https, www/non-www and trailing slash variants | The version you chose | A 301 redirect, preferable because it also unifies visitors |
| Content syndicated or republished on another site | The original article, cross-domain | A link to the source if the partner refuses the canonical |
| Printable version, AMP, separate mobile page | The main page | Removing the duplicated version |
| Product pages reachable through several categories | One unique URL per product | Restructuring the URLs so only one address is generated |
The canonical is not suited to two pages with different content: Google then ignores the signal. To merge two distinct articles, the 301 redirect is the right tool.
What Google really does with it
Google treats the canonical as a strong but non-binding signal. It combines it with other clues (redirects, internal links, sitemap, HTTPS, hreflang, the relative quality of the pages) to elect a "Google-selected canonical", which may differ from the one you declared. Search Console shows both in the URL inspection tool and flags pages as "Duplicate, Google chose different canonical than user". That case almost always reveals an inconsistency between your signals: heavy internal linking to the non-canonical version, a contrary redirect, or a canonical pointing to an empty page.
The most frequent mistakes
| Mistake | Effect |
|---|---|
| Canonical to the home page on every page (a default CMS setting) | Gradual deindexing of the entire site |
| Canonical to a redirected or 404 URL | Signal ignored, duplicates kept |
| A chain of canonicals (A to B, B to C) | Google follows only one level; A stays duplicated |
| A relative canonical, or one with a typo in the domain | Points to a non-existent URL; ignored or, worse, applied to another site |
| Pagination with a canonical to page 1 | Pages 2 onwards are no longer crawled; the products they list lose their links |
A crawl with Screaming Frog lists every canonical on the site, its target and its response code within minutes. Checking that report is part of any SEO audit, alongside the meta robots tag.
How GreenRed helps
Rather than juggling several tools, GreenRed's Site health module brings these metrics together in a single dashboard, compares them over time and tells you which actions come first. You can try it free, with no card, from the Pricing.
Frequently asked questions
Should every page have a canonical tag?
Yes — a self-referencing canonical on every indexable page is the recommended practice. It protects you against URL parameters added by third-party tools and against unintentional copies. Most CMSs and SEO plugins add it automatically; check that it points to the clean HTTPS URL.
Canonical or 301 redirect: which should you choose?
The 301 redirect when the duplicate page has no reason to exist for visitors (an old URL, an http version). The canonical when both URLs must stay reachable (filters, sorting, tracking parameters) but only one should be indexed. The 301 is the stronger of the two signals.
Can Google ignore my canonical tag?
Yes. The canonical is a hint, not a directive. If your other signals (internal links, sitemap, redirects, content) point to another URL, Google may choose that one instead. Search Console then reports that Google chose a canonical different from the one declared.
Does the canonical pass PageRank?
Yes, in the vast majority of cases. When Google accepts the canonical, it consolidates the signals — links included — on the reference URL, just as a 301 redirect would. That is precisely what makes it useful for syndicated content and for product pages reachable by several paths.