Canonicalization is the process of choosing one URL to represent a piece of content when the same or very similar content can be reached at more than one address. That chosen address is the canonical URL. It matters for SEO because search engines try to show only one version of duplicate content, and you want it to be the one you prefer.
This guide explains why duplicate URLs happen, how Google picks a canonical, which methods you can use to state your preference, how to check what Google chose and which mistakes to avoid. It uses Google’s own documentation and does not quote benchmarks, because they vary too much between sites to be useful.
Pick one URL for each page, redirect or point every other version to it, and keep your signals consistent. Google’s page on URL canonicalization says your preference is a hint, not a rule, so Google may choose a different page. The more clearly your redirects, canonical tags and sitemaps agree, the more likely it is to choose the one you want.
Table of contents
What canonicalization is
Many pages can be reached by more than one address. Google groups pages that seem the same, or that have very similar main content, into a cluster. It then picks the page that is, in its words, the most complete and useful for search users, and marks it as canonical. People often call this deduplication.
The canonical page is the one Google crawls most often and usually shows in search results. The others are crawled less often. The system exists so searchers do not see the same page ten times, and so ranking signals are not spread across copies.
Why duplicate URLs happen
Most duplicates are accidents of how sites are built, not deliberate copying. These are the common causes.
| Cause | Example | Typical fix |
|---|---|---|
| HTTP and HTTPS | http://example.com/shoes and https://example.com/shoes | Redirect HTTP to HTTPS |
| With and without www | www.example.com and example.com | Redirect one to the other |
| Trailing slash and case | /shoes and /shoes/ | Choose one form and redirect |
| URL parameters | /shoes?sort=price or ?utm_source=email | Canonical tag to the clean URL |
| Filtered or sorted lists | /shoes?colour=red | Canonical tag, or block the options you do not need |
| Print or mobile versions | /shoes/print or m.example.com/shoes | Canonical tag, or one responsive page |
| Copied or syndicated content | The same article on two sites | Canonical tag to the original, where the other site agrees |
A site with a few hundred pages can easily have thousands of URL variations. That is why canonicalization is usually a technical SEO task rather than a content task. Our technical SEO service covers this kind of check. If you use subdomains, see our guide to subdomain SEO.
How Google chooses a canonical
Google looks at several signals together. Its documentation lists these factors: whether the page is served over HTTP or HTTPS, redirects, whether the URL is in a sitemap and the rel=”canonical” link annotation. Our guide to metadata SEO covers the other tags a page needs.
Two points are worth remembering. First, the canonical you declare is a hint, not a rule, so Google can disagree. Second, search results usually point to the canonical, but a duplicate can be shown when it suits the user better. For example, a mobile version may be shown to mobile users.
Language versions are treated as duplicates only if their main content is in the same language. A French page and an English page of the same product are not duplicates of each other. Our guide to global SEO covers how to label those versions.
The methods and how strong they are
Google’s guidance on consolidating duplicate URLs lists these ways to state a preference, from strongest to weakest.
| Method | What Google says | Use it when |
|---|---|---|
| Redirect | A strong signal that the target should become canonical | You are retiring the duplicate page for good |
| rel=”canonical” link element | A strong signal that the specified URL should become canonical | Both pages must stay reachable |
| rel=”canonical” HTTP header | Works for files that are not HTML, such as PDFs | You cannot edit the page itself |
| Sitemap | A weak signal | Always, as support for the other signals |
A redirect or a canonical tag?
If people no longer need the duplicate, redirect it. If both versions must stay live, such as a page with tracking parameters, use a canonical tag. The signals can stack and become more effective when combined, as long as they agree.
Best practices
- Use full addresses. Google advises absolute URLs in canonical tags, because relative paths can cause problems over time.
- Name one canonical per page. Do not give different canonicals for the same page in different places.
- Pick the tag or the header, not both. Google says using both at once is more error prone.
- Add a self-referencing canonical. Include a rel=”canonical” link on the canonical page itself.
- Prefer HTTPS. Google prefers HTTPS over an equivalent HTTP page, unless the certificate is invalid or a redirect sends people from HTTPS back to HTTP.
- List only canonical URLs in your sitemap. It is a weak signal, but it should agree with the others.
- Keep language tags separate. Annotations for alternate versions, such as hreflang, are not used for canonicalization.
How to check what Google chose
Google’s URL Inspection tool shows two fields for an indexed page: the user-declared canonical, which is the one your page states, and the Google-selected canonical, which is the page Google chose when it found similar pages. If the two differ, Google disagrees with you.
The Page indexing report in Search Console also lists canonical statuses. Three are common.
- Duplicate without user-selected canonical. The page duplicates another one and did not declare a preference. Google chose another page. This is not an error.
- Duplicate, Google chose different canonical than user. You declared one URL, but Google indexed a different one. Look for conflicting signals.
- Alternate page with proper canonical tag. The page correctly points to the canonical, which is indexed. No action is needed.
Common mistakes
- Using robots.txt to canonicalize. Google says not to. Blocked URLs can still be indexed without their content.
- Using noindex to choose a canonical. Google does not recommend it within one site, because it blocks the page from search completely.
- Conflicting signals. A canonical tag that points one way while redirects, internal links and the sitemap point another confuses the result.
- Relative paths. They can break when the site structure changes.
- Several canonical tags. Plugins and themes can each add one. Check the final page source.
- Linking internally to non-canonical URLs. Link to the version you want to rank.
A worked example
This is a made up example, used to show the method. It is not a client result, and the numbers are not real.
Imagine an online shop whose red shoes page can be reached at four addresses: with and without www, and with and without a tracking parameter from an email campaign.
| Address | What the shop does |
|---|---|
| http://example.com/red-shoes | Redirect to the HTTPS version |
| https://www.example.com/red-shoes | Redirect to the version without www |
| https://example.com/red-shoes?utm_source=email | Stays reachable, with a canonical tag to the clean URL |
| https://example.com/red-shoes | The canonical, with a self-referencing tag, listed in the sitemap |
Internal links, the sitemap and the tags all point to the same address, so Google has no reason to pick another.
Canonicalization FAQ
What is canonicalization in SEO?
It is choosing one preferred URL for content that can be reached at several addresses, and telling search engines which one it is.
What is the difference between normalization and canonicalization?
Normalization is tidying a URL into a standard form, such as using lowercase letters and removing a default port. Canonicalization is choosing which of several different URLs for the same content should be the representative one. Normalization is often a step inside canonicalization.
Can you give an example of a canonical URL?
If a page is reachable at example.com/shoes and example.com/shoes?sort=price, the canonical is usually example.com/shoes. The tag on the sorted page names it as the preferred address.
What is a canonical error in SEO?
There is no single error. It usually means a conflict, such as Google choosing a different canonical than you declared. Search Console’s Page indexing report shows which pages are affected, and some statuses, such as duplicate without user-selected canonical, are not errors at all.
Does a canonical tag guarantee Google will follow it?
No. Google treats it as a strong signal but not a rule. Keep your redirects, internal links and sitemap consistent with it.
Where to start
Open Search Console and look at the Page indexing report. Check any page marked “Google chose different canonical than user”, compare your tag, redirects and sitemap for that page, and fix the conflict. That is a solid first month.
If you want a second opinion, Torplix offers a free audit of your site, with a written list of fixes in priority order. It is yours to keep whether or not you work with us, and we do not promise rankings. See our guide to SEO and our guide to SEO analytics for how we work.



