Canonical URL
A canonical URL tells engines which version of duplicate pages is the real one. Learn how rel=canonical works and the mistakes that hide pages entirely.
TL;DR
A canonical URL is the version of a page you want engines to treat as authoritative when several URLs serve similar or identical content. You declare it with a rel=canonical link element in the page head. Duplication is normal on real sites: tracking parameters, filter combinations, print views and protocol variants all produce multiple URLs for one piece of content.
Why canonical URLs matter
When engines see the same content at several addresses, they pick one to represent it. A canonical tag means you make that choice rather than leaving it to inference.
- Signal consolidation: Links and authority pointing at duplicate versions accrue to the page you nominated.
- Duplicate resolution: Parameterised and filtered URLs stop competing with the page they derive from.
- Crawl efficiency: Crawlers spend less time fetching near-identical variants.
- Citation consistency: AI engines that cite a source use the canonical address, keeping attribution pointed at one URL.
How to implement canonical URLs
Use absolute URLs
Include the protocol and domain. Relative canonicals are permitted and frequently resolved wrongly.
Self-canonical every page
Each canonical page should point at itself. This resolves parameter variants automatically without extra rules.
Point duplicates at the original
Filtered, sorted and tracked versions should all canonical to the clean base URL.
Keep canonicals and sitemaps consistent
Only canonical URLs belong in your sitemap. Listing a page that canonicals elsewhere sends contradictory signals.
Avoid chains
A canonical pointing at a page that canonicals somewhere else weakens the signal. Point directly at the final destination.
Check the rendered head
Canonicals injected by JavaScript are missed by crawlers that do not execute scripts, which includes most AI crawlers.
Canonical URL vs. 301 redirect
Canonical URL: A hint. Both URLs stay reachable, users can visit either, and engines are asked to credit one.
301 redirect: An instruction. The old URL stops serving content and everyone, human or bot, lands on the new one.
Use a canonical when both versions need to exist, such as a product reachable through several category paths. Use a redirect when the old URL should be gone. Engines treat redirects as far stronger signals, because they leave no ambiguity.
A misconfigured canonical can remove a page from consideration entirely while every dashboard reports the site as healthy. Findrix runs a 31-point technical audit covering canonicals, crawler access and rendering, then tracks how often seven AI engines cite you against named competitors. Every gap comes with the fix already written: technical, content and off-site. The audit is free, takes about a minute, and requires no signup.
Common canonical mistakes
- Canonicalising everything to the homepage: A catastrophic template error that tells engines no other page deserves indexing.
- Cross-domain confusion: Pointing at a syndication partner rather than your own page hands them the credit.
- Canonical plus noindex: Contradictory instructions on one page, resolved unpredictably.
- Paginated pages canonicalising to page one: Content on later pages becomes invisible.
- JavaScript-injected canonicals: Invisible to crawlers that do not render scripts.
The failure that looks like nothing is wrong
Canonical errors are unusually dangerous because they produce no visible symptom. The page loads normally, returns a 200 status, appears in your sitemap and passes every uptime check. It simply stops being eligible for indexing, and nothing anywhere reports that.
The template-wide version is worse. A developer hardcodes a canonical during a redesign, it ships across every page, and the site tells engines that one URL represents the entire domain. Traffic decays over weeks rather than collapsing overnight, which delays diagnosis further.
Check the canonical in the raw HTML source of each page template after every release, and confirm it matches the URL you want people to land on.
Frequently asked questions
Is a canonical tag a directive or a hint?
A hint. Engines usually respect it but can override it when other signals disagree, such as internal links consistently pointing elsewhere or the canonical target being substantially different content. A 301 redirect is a directive.
Should every page have a self-referencing canonical?
It is good practice. Self-canonicalisation resolves parameter and tracking variants automatically, costs nothing, and removes ambiguity about which version of a URL is authoritative.
Can I canonicalise to a different domain?
Yes, and it is used legitimately in syndication, where a republished article canonicals back to the original. Do it accidentally and you have told engines that another domain deserves credit for your content.