Canonicals and redirects both consolidate two URLs into one, and the difference is whether the original stays reachable. A redirect removes it; a canonical keeps it available while telling search engines which version to index.
Choosing between them is usually straightforward once framed as a user question rather than a technical one: does anyone still need to visit the original URL?
The deciding question
- Nobody needs the original: redirect. It is unambiguous and obeyed rather than interpreted.
- Users still need it: canonical. The page stays usable while signals consolidate elsewhere.
That covers most cases. The remaining nuance is that a redirect is a directive and a canonical is a hint, so a redirect is stronger where you have a genuine choice.
Redirect cases
- URLs retired during a restructure or migration.
- Two pages genuinely merged, where only the merged version should exist.
- Protocol and www variants, which nobody needs to reach separately.
- Trailing slash and letter-case variants, which should resolve to one form at server level.
- Discontinued products with a replacement. See discontinued products.
Canonical cases
- Filtered or sorted category views that users navigate to deliberately.
- Product variant parameter URLs, where the visitor selected the variant.
- Tracking and campaign parameters, which must keep working.
- Print or export views.
- Products reachable through several category paths, all of which should still function.
Where people get it wrong
Canonicalising when they should redirect
Leaving a retired URL reachable with a canonical means it can still be linked, crawled and occasionally indexed despite the declaration. If nobody needs it, a redirect is cleaner and stronger.
Redirecting when they should canonicalise
Redirecting a filtered view breaks the feature for the user. Someone selecting a filter and being thrown to the unfiltered category has had something taken away.
Doing both
A page that redirects cannot serve a canonical, since it never renders. Where both are configured, the redirect wins and the canonical is irrelevant.
Canonicalising to a redirecting URL
The canonical target should return 200. Pointing at a URL that redirects breaks the chain and forces search engines to work out your intent.
The third option: noindex
Neither consolidates. Noindex removes a page from results without transferring anything to another URL, which is right for pages that have no equivalent, such as internal search results. Applying it to something that should have been canonicalised throws value away. See canonical vs noindex.
Frequently asked questions
Which is stronger?
A redirect. It is a directive and is obeyed; a canonical is a hint that can be overridden when other signals disagree. Where both would work and the original is not needed, redirect.
Can I canonicalise across domains?
Yes, cross-domain canonicals are supported and are appropriate for syndicated content. For a permanent move of your own site, a redirect is the better instrument.
Why is my canonical being ignored?
Usually contradicting signals: internal links, sitemap entries or redirects pointing elsewhere. See Google chose a different canonical.
Does a canonical pass link equity?
It consolidates signals onto the target much as a redirect does, when honoured. The difference is reliability rather than mechanism.
Consulting CTA
If you are unsure which of your duplicate URLs should redirect and which should canonicalise, book an SEO consultation for a URL consolidation review.