What is a canonical tag?

What rel=canonical does

A canonical tag is a link element in the head of a page that names the preferred URL when the same or very similar content is available at more than one address. It asks search engines to treat that URL as the main version, show it in results, and consolidate signals such as links onto it.

It is a strong hint, not a directive. Google weighs it alongside other signals, including redirects, internal links, sitemap entries and whether a URL uses HTTPS, and can choose a different canonical when the signals disagree. The URL Inspection tool in Search Console shows both the canonical you declared and the one Google selected.

Where duplicate URLs come from

Most duplication is accidental. The same page often answers at several addresses:

• Tracking, sorting and filter parameters, such as ?utm_source=newsletter or ?sort=price.
• HTTP and HTTPS versions of the same page.
• www and non-www hostnames.
• With and without a trailing slash.
• Print-friendly versions, and products that sit under more than one category path.

For protocol, hostname and trailing slash variants, a site-wide 301 redirect is the stronger fix. Use canonical tags when both URLs need to keep working, as with parameter variants.

How to set it correctly

Put one canonical link in the head of the HTML. Use an absolute URL, including the protocol and hostname, and point it at a URL that returns a 200 status and is indexable.

<link rel="canonical" href="https://www.example.com/shoes/trail-runner/">

Add a self-referencing canonical to every indexable page, so each one names itself. Generate it from the clean URL rather than the requested one, so parameter and tracking variants point back to the main version. For PDFs and other non-HTML files, send the same hint in a Link HTTP header.

Common mistakes

• Pointing the canonical at a URL that redirects, returns a 404 or carries noindex. Point it at the final, indexable URL.
• Setting every paginated page to canonicalize to page 1. Google advises giving each page in a series its own canonical, because later pages list different items.
• Sending conflicting signals: the canonical names one URL while the sitemap, internal links or hreflang tags use another. Keep them all pointing at the same version.
• Declaring more than one canonical, or injecting one with JavaScript that differs from the one in the HTML. Google may ignore conflicting canonicals.
• Placing the tag in the body. Google only accepts rel=canonical in the head.
• Using noindex or robots.txt to handle duplicates. Neither consolidates signals, and Google advises against both for choosing a canonical.