What is a canonical URL?
A canonical URL is the address selected as the main version of a page within a group of duplicate or near-duplicate URLs. Canonicalisation is the process of grouping those similar pages and choosing the representative version.
For example, one product might be available at several addresses:
https://www.example.com/shoes/red-trainer/
https://www.example.com/shoes/red-trainer/?colour=red
https://www.example.com/shoes/red-trainer/?utm_source=emailIf the content is effectively the same, the clean product URL may be the best canonical. Search engines can then focus their indexing and ranking systems on one representative address rather than treating every parameter variation as a separate page.
The canonical you declare is a preference, not an absolute instruction. Google evaluates redirects, canonical tags, sitemap inclusion, internal links, content similarity and other signals before selecting its canonical URL. It can choose a different version when the evidence points elsewhere.
Why canonical URLs matter
Duplicate URL versions are common. They can arise from tracking parameters, filters, sorting, print pages, HTTP and HTTPS versions, www and non-www hosts, session identifiers and copied product paths.
A clear canonical setup helps search engines:
- Identify the URL you want shown in search.
- Consolidate signals associated with duplicate versions.
- Spend less time crawling unnecessary variations.
- Understand which page should be evaluated as the main version.
- Keep search reporting focused on a stable address.
Canonicalisation does not make unrelated pages rank better, and it does not hide confidential content. Duplicate pages may still be crawled. Use access control for private material and a noindex directive when an accessible page must stay out of search for a reason other than duplicate consolidation.
What does a canonical tag look like?
For an HTML page, place a rel="canonical" link element in the document’s <head>:
<link rel="canonical" href="https://www.example.com/shoes/red-trainer/">Use a complete absolute URL, including https:// and the preferred hostname. The tag should appear once and point to a working, indexable page that contains the same or very similar main content.
The preferred page can include a self-referencing canonical:
<link rel="canonical" href="https://www.example.com/guides/technical-seo/">Self-referencing canonicals make the preferred version explicit and help protect against unplanned parameter or tracking variants.
For non-HTML files such as PDFs, a canonical can be sent in the HTTP response header:
Link: <https://www.example.com/preferred-document/>; rel="canonical"Search engines can also infer canonical preference from permanent redirects and sitemap inclusion. Google describes redirects and rel="canonical" as stronger signals, while sitemap inclusion is weaker. Consistent signals can reinforce one another.
When should you use a canonical tag?
Use a canonical when several accessible URLs need to remain available but represent the same or substantially similar content.
Common cases include:
- Tracking parameters added to an otherwise identical landing page.
- Product URLs reached through several category paths.
- Sort or filter URLs that do not create a meaningfully different page.
- Print-friendly versions of an article.
- Syndicated or duplicated pages you control, after checking the appropriate cross-domain approach.
- An A/B test with separate URLs for close variants.
- Protocol or hostname variations during a period when duplicates remain accessible.
Before applying the tag, compare the main content. The canonical target should be a genuine substitute for the duplicate page. If a visitor landing on the target would lose essential information or reach a different purpose, the two pages may not belong in the same canonical group.
When is a canonical tag the wrong tool?
When a page has moved permanently
Use a permanent server-side redirect when the old URL no longer needs to remain available. A redirect moves visitors as well as search engines, while a canonical leaves both URLs accessible.
When a page must stay out of search
Use noindex for an accessible page that should not appear in search and is not simply a duplicate being consolidated. Use authentication for private content. A canonical does not guarantee exclusion of the source URL.
When the pages serve different purposes
Do not canonicalise every similar product, location or article to one page merely because they share a template. If each page gives visitors distinct and valuable information, it may deserve a self-referencing canonical.
When the target is broken or excluded
Do not point a canonical at a redirect, error page, noindexed URL or blocked resource. The target should be the final, working, indexable version you want search engines to select.
For paginated sequences
Pages two, three and beyond usually contain different items. Canonicalising every page in a sequence to page one can hide that distinction. Give each useful page its own canonical unless it is genuinely a duplicate.
Check canonical URLs in Clearcrawl
Clearcrawl shows the canonical declared on each live page and helps expose patterns across templates. A single mistake may affect one URL; a CMS rule can point an entire section at the wrong page.
Use the crawl to understand what your site declares. Then use Google Search Console’s URL Inspection tool to compare the user-declared canonical with the Google-selected canonical for important pages.
Canonical URL examples
Tracking parameters
An email campaign adds a tracking parameter:
Duplicate: https://www.example.com/pricing/?utm_source=newsletter
Canonical: https://www.example.com/pricing/The tracked URL can remain measurable while the clean URL is declared as the preferred search version.
Product category paths
The same product appears through two category routes:
https://www.example.com/mens/trainers/red-runner/
https://www.example.com/sale/red-runner/Choose one stable product URL as canonical and make internal product links use it consistently.
Filtered category pages
A category can be sorted in several ways:
https://www.example.com/chairs/?sort=price
https://www.example.com/chairs/?sort=ratingIf sorting changes only the order, both may canonicalise to the main /chairs/ category. If a filter creates a valuable landing page with distinct demand and content, it may deserve its own canonical URL instead.
HTTP and HTTPS
The HTTPS version should normally be the stable destination:
http://www.example.com/about/
https://www.example.com/about/Use permanent redirects from HTTP to HTTPS and make canonicals, sitemaps and internal links reference HTTPS. The redirect is the main action because the HTTP page no longer needs to remain available. Canonical signals matter most when pages overlap; compare duplicate content patterns with Google’s indexing decisions.
How to audit and fix canonical issues
- 01Choose the preferred version.
Select the URL that offers the clearest, most stable experience. Prefer HTTPS, a consistent hostname and a clean path. Confirm that it is the page you would want a visitor to see in search.
- 02Compare the content.
Make sure each source page is duplicate or substantially similar to the target. If the pages answer different needs, give them separate canonicals and strengthen their distinct content.
- 03Check the canonical implementation.
Confirm that the HTML contains one canonical link in the
<head>, or that the appropriate HTTP header is used for a non-HTML resource. Use an absolute URL and check the rendered page when JavaScript can alter the head. - 04Test the target.
The canonical destination should resolve directly to
200 OK, allow indexing and declare itself as canonical. Avoid chains in which the declared URL redirects somewhere else. - 05Align every supporting signal.
Use the preferred URL in:
- Internal links.
- XML sitemaps.
- Hreflang annotations where applicable.
- Structured data URLs.
- Redirects from retired duplicate versions.
Do not list one URL in the sitemap while declaring another as canonical. Mixed signals make the intended result less clear.
- 06Fix the source of repeated errors.
Trace a widespread problem to the CMS, ecommerce platform, plugin or page template. Correcting the generator prevents the same conflict from returning whenever a page is published.
How to verify the preferred URL
Start with a fresh crawl:
- Confirm that the duplicate pages point to the intended target.
- Confirm that the target has a self-referencing canonical.
- Check that the target is working and indexable.
- Make sure internal links and the sitemap use the preferred URL.
- Look for multiple or conflicting declarations.
Then inspect representative URLs in Google Search Console. Compare the user-declared canonical with Google’s selected canonical. If they differ, review the content and all supporting signals before assuming Google is wrong.
After a meaningful repair, request re-indexing for a small number of important examples. Canonical selection occurs during indexing, so a successful live test cannot guarantee which URL Google will ultimately choose. Allow time for the pages to be recrawled and reassessed.
Common mistakes
The homepage is not a substitute for unrelated products, categories or articles. Canonical targets need substantially similar content.
Declare the final working URL. A redirecting, missing or noindexed target weakens the signal and creates an avoidable conflict.
A canonical tag, sitemap and internal links should not nominate different versions. Align them around one preferred address.
Multiple declarations can be ignored. Keep one clear canonical instruction in the appropriate location.
Robots.txt controls crawling, and noindex removes a page from consideration. Use canonical signals to consolidate duplicate pages and reserve the other controls for their intended purposes.
Frequently asked questions
What is a canonical URL in SEO?
It is the representative URL chosen for a group of duplicate or very similar pages. Search engines can consolidate their understanding and usually show the canonical version in search results.
Is a canonical tag a directive?
No. It is a strong signal, but Google can select another canonical when content or other site signals support a different URL. Align redirects, internal links and sitemaps with the tag.
Should every page have a self-referencing canonical?
It is generally useful for indexable canonical pages because it makes the preferred address explicit. The tag still needs to be generated accurately and must not hide separate pages that serve different purposes.
Can a canonical URL point to another domain?
Cross-domain canonicals can be used for duplicate content in suitable cases, but the target must be the genuine preferred version and both parties should understand the effect. Google may still choose another canonical.
What is the difference between a canonical tag and a redirect?
A redirect sends visitors and crawlers to another URL, so use it when the old page no longer needs to remain accessible. A canonical keeps both URLs available while signalling which similar version should be treated as representative.
ClearcrawlMake every preferred URL clear
Crawl your website to uncover missing tags, broken targets, redirects, duplicate declarations and mixed sitemap signals. Review the affected templates, align each preferred URL and crawl again to confirm the correction.
7 free days. See what Clearcrawl can find, then choose the plan that fits your site.