SEO

Canonicalization

Also called URL canonicalisation

The process of settling on one preferred address when the same content is reachable at several URLs.

Quick facts: Canonicalization

Category
SEO
Also called
URL canonicalisation
Level
Intermediate
Affects
Duplicate handling, index selection, reporting accuracy
Where to see it
Search Console URL Inspection (Google-selected canonical), Screaming Frog, page source
In this article4
  1. How canonicalization works
  2. Why canonicalization matters
  3. Where canonicalization goes wrong
  4. How to get it right

How canonicalization works

The same page can usually be reached at more addresses than anyone realises: with and without www, over HTTP and HTTPS, with and without a trailing slash, in different letter cases, with a tracking parameter attached, with a session identifier, through a filter combination, or as a print version. To a person these are one page. To a search engine they are separate addresses that happen to look alike.

Canonicalization is the process of resolving that group down to one representative address. Search engines do it whether you help or not, weighing several signals together: the canonical link you declare, where your redirects lead, which version your internal links use, which one is listed in the sitemap, whether the page is served securely, and how the addresses relate to each other. Your declared canonical is treated as a strong hint in that decision rather than a command, which is why contradicting yourself elsewhere can override it.

Why canonicalization matters

Signals that should sit on one page get divided between copies. Links from other sites arrive at whichever version was pasted, engagement is split, and no single address ends up as strong as the content deserves. On an online shop, where filters and sorting can generate an enormous number of addresses for the same products, this is the difference between a category page that competes and one that never quite does.

There is a reporting cost too. When traffic is spread across several addresses, no report shows the true picture of a page, and comparisons across time break as soon as the split changes.

Where canonicalization goes wrong

Contradiction is the usual failure. The canonical tag names one address, the sitemap lists a second, internal links point at a third, and a redirect sends visitors to a fourth. Faced with four opinions the search engine picks its own, and it may not pick yours.

Pointing every page at the homepage is another, usually a plugin misconfiguration, and it removes everything else from consideration. Using a canonical where a redirect belongs is a third: if an address should not exist at all, redirect it rather than leaving it reachable. And treating the tag as a guarantee leads people to stop checking, when the only reliable answer is which address the search engine actually chose.

How to get it right

Decide your one true form of address — scheme, host and trailing slash — and enforce it with redirects at a single layer, so no other rule can disagree. Give every indexable page a canonical pointing at itself in that exact form.

Then make everything else agree. Internal links, navigation, the XML sitemap, hreflang annotations and any campaign landing pages should all use the chosen address. Verify with the URL Inspection tool, which reports the address you declared alongside the one the search engine selected; a mismatch between those two is your list of things to fix. Where near-identical pages exist for a real reason, decide whether the honest answer is a canonical, a redirect, or merging them into one page — a judgement that belongs in a proper SEO audit rather than in a plugin setting, and one that overlaps with duplicate content decisions.

Do and do not

Do

  • Fix one scheme, host and slash, then enforce it
  • Give every indexable page a self-referencing canonical
  • Make sitemap, internal links and canonicals agree

Do not

  • Point unrelated pages at the homepage
  • Use a canonical where a redirect belongs
  • Assume the tag you set was the one chosen

Questions people ask about this

Is a canonical tag a rule that Google must follow?

No, it is a strong hint. Google weighs it alongside your redirects, internal links, sitemap entries and which version is served securely, then selects one address to represent the group. When your own signals disagree with each other, the tag can lose. Keeping everything consistent is what makes the declared address the selected one.

Should I use a canonical or a redirect?

Redirect when the address should not be reachable at all, such as an old URL after a restructure or a duplicate host. Use a canonical when the address needs to stay usable for visitors but should not be indexed separately, such as a URL carrying a tracking parameter or a filtered version of a category page.

Does every page need a canonical tag pointing at itself?

It is good practice. A self-referencing canonical states clearly which form of the address you consider correct, which helps when a page is reached with a parameter attached or in a different letter case. It costs nothing, prevents a common class of accidental duplication, and makes any later mistake far easier to spot.

Related terms

Found this useful?

Share it, or ask an AI to summarise it

Back to the glossary

Knowing the term is the easy part

Applying it to your own site and budget is the work. Book a call and I will tell you what actually applies to you.