What is a canonical URL?

A canonical URL is an HTML tag that tells search engines which version of a page is the "original" and should be indexed. It's used when the same or very similar content is available at multiple URLs — a situation far more common than one might think.

What does the canonical tag look like?

<link rel="canonical" href="https://example.com/article" />

This tag is placed in the <head> section of an HTML page. It tells Google: "this is the preferred URL for this content — index this one, not the other variants."

The canonical URL can also be specified in the HTTP response header (Link: <URL>; rel="canonical") — useful for PDFs and other resources without a <head> section.

When to use canonical?

A canonical URL is essential wherever the same content is available at multiple addresses. The most common scenarios:

URL parameters — /products?sort=price and /products?sort=name have the same content, differing only in sorting
www/non-www versions — www.example.com vs example.com
HTTP/HTTPS — http:// vs https:// (although this should be resolved with a 301 redirect)
Trailing slash — /article vs /article/
Content syndication — when the same article appears on multiple sites (e.g., a reprint on an industry portal)
Mobile versions — m.example.com vs example.com (less common with responsive design)
Pagination — /blog?page=1, /blog?page=2 may have a canonical pointing to the main category page
Product variants — the same product in different colors under different URLs

Why is canonical important?

Without a canonical URL, Google has to decide on its own which page variant to index — and it often chooses incorrectly. The consequences of missing canonicalization can be serious — which is why it's a key item in every SEO audit:

Duplicate content issues

Authority dilution — backlinks pointing to different URL variants don't aggregate; instead of one strong page, you have several weak ones
SERP cannibalization — multiple versions of the same content compete against each other in search results
Wasting crawl budget — Google wastes time crawling duplicates instead of discovering new content
Inaccurate analytics — traffic split across multiple URLs makes analysis difficult

Benefits of proper canonicalization

Authority consolidation — all ranking signals go to a single, preferred URL
Clean data in Google Search Console — one URL per piece of content, easy analysis
Control over the index — you decide what Google indexes
Better crawlability — the crawler uses the budget more efficiently

How to properly implement canonical?

Implementation rules

Every page should have a canonical — even if it points to itself (self-referencing canonical). You can find the full list of elements to verify in the technical SEO checklist
Canonical must point to the HTTPS version — always use a full, absolute URL
Canonical should point to a 200 page — not to a redirect or a 404 page
One canonical per page — multiple canonical tags on a single page are ignored
Consistency with other signals — canonical should align with sitemap, hreflangs, and internal linking

Common mistakes

Canonical to a noindex page — conflicting signals; Google may ignore both
Canonical chains — A -> B -> C; Google may not follow the chain
Canonical to a page with different content — Google ignores the canonical if the content significantly differs
Missing canonical on pagination pages — leads to indexing of hundreds of /page/2, /page/3 pages

Canonical URL: Relative or Absolute?

Use an absolute URL. A relative canonical such as /article is valid HTML and Google can process it, but the browser and the crawler resolve it against whatever protocol and host served the page. On an http:// copy, a www variant, or a staging server, it points to that copy instead of the version you want indexed.

Google's own documentation recommends absolute URLs for rel="canonical" for the same reason. Two details are easy to miss:

A path without a leading slash (article) resolves relative to the current directory. On /blog/article it points to /blog/article, but if the same page is also reachable at /archive/article, there it points to /archive/article — each copy names itself as the original.
Templates that build the canonical from the request inherit the same problem as a relative URL. The canonical should come from one configured base URL, not from the host name of the current request.

Canonical vs 301 redirect

Canonical is a "suggestion" for Google — the search engine may ignore it. A 301 redirect is a "directive" — the browser and crawler are physically redirected. When possible, prefer a 301 redirect over canonical — it's a stronger signal.

Situation	Solution
One piece of content, one correct URL, the rest to be removed	301 redirect
One piece of content, multiple URLs must exist (parameters, sorting)	Canonical
Content syndication on an external site	Canonical
Domain migration	301 redirect + canonical

Canonical Domain: Choosing One Host Name

The canonical domain is the single combination of protocol and host name you want search engines to index — for example https://example.com, not http://example.com or https://www.example.com. Choose one version, redirect every other version to it with a 301, and use it consistently in canonical tags, the sitemap, and internal links.

www or non-www. Neither version ranks better on its own. What matters is that only one of them serves content and the other returns a 301 for every path, not just the homepage.
HTTP and HTTPS. HTTPS should be the canonical protocol. A canonical tag on an HTTP page that points to HTTPS helps, but the HTTP version should also redirect.
Several domains with the same content. When an old domain, a misspelled domain, or a campaign domain serves a copy of the site, redirect it to the canonical domain. A cross-domain canonical tag is a weaker substitute for cases where the copy has to stay online.
Checking what Google chose. The URL Inspection tool in Google Search Console shows both the canonical you declared and the one Google selected. When they differ, look for signals that contradict each other, such as redirects, sitemap entries, or internal links pointing to different versions.

Related terms

Duplicate content — the problem that canonical solves
301 Redirect — permanent redirect, an alternative to canonical
Crawlability — the ability of crawlers to traverse a website
Indexing — the process of recording a page in Google's index
Crawl budget — crawl budget, wasted by duplicates

Canonical URL