← Back to Insights

Canonical Tag vs Redirect: Which Should You Use?

Compare canonical tags and redirects, including how each method affects visitors, crawlers, duplicate pages and consolidated search signals.

Canonical Tag vs Redirect: Which Should You Use?

Canonical tags and redirects can both consolidate signals around a preferred URL, but they change the user experience in different ways. A redirect sends the visitor and crawler to another address. A canonical tag leaves the current page accessible while suggesting that a different URL should represent the duplicate set in search.

The right choice depends on whether the old or alternate URL still needs to work as an independent destination. Choosing based only on perceived SEO strength can create confusing navigation and unnecessary crawl paths.

The simplest distinction

Use a permanent redirect when the original URL should no longer be used. Use a canonical when multiple accessible URLs need to remain available but contain the same or nearly the same main content.

Google recommends permanent server-side redirects, such as 301 or 308, when a page has moved. Its redirect documentation treats them as the clearest way to direct users and Search to the new location.

When a redirect is usually correct

  • A page slug changes permanently.
  • HTTP URLs move to HTTPS.
  • www and non-www hosts are consolidated.
  • Two duplicate pages are merged into one resource.
  • A site migration changes the URL structure.
  • An obsolete page has a close replacement that satisfies the same intent.

Update internal links and sitemap entries to point directly to the destination. The redirect remains useful for old bookmarks, backlinks, and cached addresses.

When a canonical is usually correct

  • Tracking parameters must remain usable for measurement.
  • A product is accessible through several category paths.
  • Print and standard views both serve users.
  • Sort orders display substantially the same collection.
  • Partner or syndicated versions remain publicly accessible.

The alternate page should contain a canonical pointing to the preferred version. The preferred page should normally reference itself.

Do not redirect unrelated expired pages

Redirecting every deleted page to the homepage is not helpful consolidation. A redirect should lead to a relevant replacement. When no equivalent exists, return 404 or 410 and provide useful navigation on the error page.

Search engines may treat irrelevant redirects as soft 404s. Users also become frustrated when a specific old link unexpectedly opens a generic homepage.

A canonical is not a navigation instruction

Visitors who open a canonicalised URL stay on that URL. If the alternate address causes confusion, duplicates analytics, breaks sharing, or should disappear from use, a redirect is the better mechanism.

This user-facing difference is why canonicals should not be used as a substitute for routine URL cleanup.

What happens when both are present?

If URL A redirects to URL B, the canonical tag on A is usually irrelevant because the crawler receives the redirect instead of the page HTML. The canonical on B should normally point to B or another final preferred URL.

A redirect chain that ends on a page canonicalised elsewhere creates avoidable complexity. Point the original URL directly to the final canonical destination whenever possible.

Temporary changes need temporary redirects

Use 302 or 307 when the move is genuinely temporary and the original URL should remain the long-term search destination. Examples include short maintenance periods and controlled experiments.

Do not leave temporary redirects in place indefinitely without reviewing them. Search engines can interpret long-standing configurations based on broader signals, and operational teams may forget the original intent.

How canonical tags fit into duplicate management

The detailed mechanics and mistakes are covered in the practical canonical tag guide. The central point is that canonicalisation is a signal. Support it with consistent internal links, sitemap URLs, hreflang annotations where applicable, and similar page content.

If the pages are not duplicates or close alternatives, a canonical may be ignored. Do not point thin or low-performing pages to a strong page merely to transfer value.

Decision examples

Campaign tracking URL

A newsletter link adds campaign parameters but displays the same article. Keep the parameter URL accessible and use a canonical to the clean article URL.

Changed article slug

An editor replaces an inaccurate slug with a permanent new one. Redirect the old URL directly to the new URL and update all internal links.

Printable version

A print-friendly page remains useful but duplicates the article. Canonicalise it to the main article, unless the implementation can avoid a separate URL entirely.

Expired event

If the next edition answers the same user intent, a redirect may be reasonable. If the old event remains historically useful, keep it live with updated context. If it has no value or replacement, return 404 or 410.

Audit after year-end URL changes

December publishing often includes annual archive updates, new year paths, and campaign retirements. Create a redirect map before changing URLs. After deployment, crawl old addresses, check final status codes, and verify that destination canonicals agree.

Also update the sitemap. It should list the final URLs rather than old addresses that redirect.

A reliable decision rule

Ask whether users should stay on the alternate URL. If no, redirect. If yes, and the content duplicates another preferred page, use a canonical. If the content is unrelated, use neither merely to force consolidation.

Clear URL behaviour produces clearer search signals. The best solution is usually the one that also makes immediate sense to a visitor.

Build authority you can keep working with.

If you are already investing in content and off-site SEO, a dedicated editorial portfolio can add a managed publishing layer around your priority pages and campaigns.

Request a portfolio assessment