What Is a Canonical Tag? 5 Rules for Using It

A canonical tag (`rel="canonical"`) is an HTML link element that tells Google which URL you want shown in search results when the same or very similar content is reachable at more than one address. Put simply, it is how you say "this is the main page" among duplicates.
Store filters, UTM parameters, http and https versions, URLs with and without a trailing slash... A site can end up with dozens of copies of the same page without anyone noticing. In this article I explain what the canonical tag does, how Google evaluates it, and the 5 rules that work best in practice. You will also find a short definition in the SEO and AI glossary.
Short answer: A canonical tag is a link element that points Google to your preferred URL among duplicate or similar pages. Google uses it as a strong signal but treats it as a hint, not a rule; if other signals conflict, it can choose a different page as canonical.
What does a canonical tag do?
Google groups pages whose content is the same or very similar and picks one of them as the canonical page. That page is usually the one shown in search results, while duplicates are crawled less often. The canonical tag lets you state your preference in that choice.
In practice it gives you three benefits: you influence which URL ranks, performance data in your reports consolidates on one address, and Googlebot spends less time crawling duplicates. On large sites that last point matters for crawl budget too.
The tag goes in the page's `<head>`. For non-HTML files such as PDFs, the same information can be sent in an HTTP header:
<!-- https://www.examplestore.com/womens-shoes?color=black&sort=price -->
<head>
<link rel="canonical" href="https://www.examplestore.com/womens-shoes" />
</head>
<!-- HTTP header for non-HTML files such as PDFs -->
Link: <https://www.examplestore.com/catalog.pdf>; rel="canonical"Does Google treat the canonical tag as a rule?
No. Google Search Central documentation states clearly that indicating a canonical preference is "a hint, not a rule". Google weighs signals such as HTTPS, redirects, sitemaps and the canonical tag together, and it may pick a different page than you did.
| Method | Signal strength per Google | When to use it |
|---|---|---|
| 301 redirect | Strong signal | When the duplicate page is being retired |
| rel="canonical" tag | Strong signal | When the duplicate must stay accessible (filters, parameters) |
| Canonical in HTTP header | Same purpose as rel="canonical" | Non-HTML files such as PDFs |
| XML sitemap | Weak signal | Listing preferred URLs on large sites |
So do not treat the canonical tag as a fix on its own. If your internal links, sitemap and redirects point to another address, the tag carries less weight.
5 rules for using the canonical tag
1. Add a self-referencing canonical to every page
Google recommends adding a canonical tag on the canonical page itself (a self-referencing canonical). That way, when someone links to your page with parameters or it opens through a campaign URL, your preferred URL stays clear.
2. Use absolute URLs and place the tag in the head
The `href` should be a full address with protocol and domain; Google recommends absolute paths over relative ones. The tag is only accepted inside `<head>`. Do not specify a URL fragment (`#section`) as canonical. On JavaScript sites, set the tag in the HTML source and do not let JavaScript change it; see JavaScript SEO for details.
3. Point parameter and filter URLs to the right page
If sorting (`?sort=price`), tracking (`?utm_source=`) or session parameters do not change the content, the canonical should point to the clean URL. But if a filter page such as "black leather boots" has its own search demand and should rank separately, give it its own canonical. You can plan which parameter gets which treatment with the scheme in URL structure and SEO.
4. Do not send conflicting signals
If the canonical tag points to URL A, the sitemap to URL B and internal links to URL C, Google is very likely to make its own choice. Google explicitly asks you not to specify different URLs for the same page in your sitemap and canonical tag. On multilingual sites the canonical should point to a page in the same language; for the hreflang setup see hreflang and international SEO and the hreflang generator.
5. Do not use noindex, robots.txt or the removal tool instead
Google says not to use robots.txt for canonicalization, does not recommend using noindex to steer canonical selection within a single site, and notes that the URL removal tool hides all versions of a URL from Search. A canonical tag on a page blocked by robots.txt cannot even be read.
Common canonical mistakes
The canonical problems I run into most during technical audits usually come from a single theme or plugin setting, which is why they affect thousands of pages at once.
- The canonical points to an indexable page returning 200
- Each page has exactly one canonical tag
- The sitemap lists only canonical URLs
- Internal links point to the canonical URL
- The canonical points to a redirected or 404 URL
- Theme and SEO plugin output two different canonicals
- Every page's canonical points to the homepage
- The canonical page also carries noindex
Most of these also appear as part of a longer list in common technical SEO mistakes. Canonicals left on the old domain during a migration are a separate risk; the site migration SEO guide covers that check step by step.
How do you check a canonical tag?
- Inspect the sourceConfirm there is a single rel="canonical" in the head and that it uses an absolute URL.
- URL InspectionIn Search Console, compare "User-declared canonical" with "Google-selected canonical".
- Page indexing reportReview the URLs under "Duplicate, Google chose different canonical than user".
- Crawl the whole siteUse a site crawler to list URLs whose canonical points to 3xx, 4xx or noindex pages.
I explain where these reports live and how to read them in the Google Search Console guide. A short checklist before you audit:
- Every page you want indexed has a self-referencing canonical
- Canonical URLs use https and your preferred domain version
- Canonicals on parameter URLs follow a deliberate rule
- Sitemap, internal links and hreflang point to the same canonical URLs
Canonical tag or 301 redirect?
If visitors never need to see the duplicate page, a 301 redirect is the stronger and clearer solution; Google itself recommends redirects only when you are retiring a duplicate. For duplicates users do need, such as filter, sort or campaign URLs, the canonical tag is the right tool.
I cover the difference in detail in 301 vs 302 redirects, and you can build the rules with the redirect generator. If you want duplicate content issues on your site handled as part of a thorough technical SEO audit, we can book a free intro call about SEO consulting.
Frequently Asked Questions
Does a canonical tag improve rankings?
Not directly. It helps consolidate signals from duplicate URLs onto one address so the right page appears. Its real effect is removing confusion.
Can I use a canonical across different domains?
Yes. When the same content is published on another site, for example a syndicated article, that site can point a canonical to your original page. It is still a hint, and it depends on the other site implementing it.
How do I add a canonical tag in WordPress?
Plugins such as Yoast SEO and Rank Math add a self-referencing canonical to every page automatically and let you change it per page. Check the page source to see whether your theme also outputs one; there should not be two.
Why does Google ignore the canonical I chose?
The usual reasons are internal links and the sitemap pointing elsewhere, the two pages actually having different content, or the canonical URL redirecting or returning an error. URL Inspection shows the address Google picked; align your signals with your preferred one.
Is a page with a canonical tag still crawled?
Yes. Duplicates are still crawled, but according to Google less often than the canonical page. So a canonical reduces crawl load but does not remove it.

8+ years of hands-on SEO and GEO experience across 150+ projects in e-commerce, SaaS, healthcare and local business sectors. Writes based on real client data, not theory.





