How to Choose Between a Canonical Link and a Redirect

When two URLs on a website serve the same or very similar content, search engines need signals to choose which URL to show in search results. A canonical link tells search engines which version you prefer while leaving the requested page available. A redirect sends visitors and crawlers from one URL to another.

Choose based on what visitors should see when they open the old or alternate URL. If it should stop serving a page and send visitors to a relevant replacement, use a permanent redirect. If it must remain accessible but should not be the preferred search result, use a canonical link. If a removed page has no relevant replacement, return a not-found response rather than redirecting to an unrelated page.

Choose based on what visitors should see

A canonical link tells search engines which URL you prefer. The alternate URL still loads its page, and the browser address does not change. A page can declare its preferred version in its HTML or in an HTTP response header, including for files such as PDFs. For example, its HTML can contain:

A redirect changes the visitor’s destination. The server responds to a request for the old URL with a redirect status and a Location header naming the new URL. The browser then requests that destination.

Use this decision rule:

For a permanent move, Google recommends a server-side 301 or 308 redirect where possible. For a temporary move, use a temporary status such as 302, 303, or 307 when that matches the situation. Google specifically recommends a temporary redirect for search-related site experiments, rather than signaling that the move is permanent. See Google’s guidance on redirects and Search and website testing.

What Google does with each signal

Google treats both permanent redirects and rel="canonical" as strong signals when choosing a canonical URL. The canonical URL is the one Google shows in search results or indexes for a group of duplicate or very similar pages. A redirect sends Google to the destination. A canonical link identifies the preferred URL while leaving the alternate page accessible. A sitemap entry is a weaker signal. Google considers these signals together, but it can select a different canonical from the one a site owner declares. Its canonicalization guidance describes the selection process and its limits.

Canonicalization can combine signals for duplicate URLs, make search performance reports easier to understand, and reduce how often Google crawls duplicate pages. Those are stated purposes, not guarantees of a ranking increase. Google also says ordinary duplicate content does not, by itself, violate its spam policies. The public documentation does not establish a controlled performance comparison showing that a correctly implemented canonical produces a particular result compared with a correctly implemented redirect.

For a canonical link, use an absolute URL and make sure the preferred page points to itself. Google supports either an HTML declaration or an HTTP header and recommends avoiding conflicting declarations across methods. Set the canonical to the URL you want to represent the page. Keep the sitemap entries and internal links consistent with that URL. Google’s instructions for specifying a canonical explain these implementation choices.

Check the response and Google’s selected URL

First, decide whether visitors should still see a page at the old URL. Then inspect what the server actually returns. For a redirect, check the status code and Location header, follow the redirect chain, and confirm the final destination loads the relevant replacement. For a canonical, inspect the page HTML and response headers to confirm the declared URL is the intended one and the preferred page loads successfully.

Check that the preferred URL is accessible to crawlers and eligible for indexing. Also check that internal links and sitemap entries point to it. A canonical declaration does not fix a destination that fails to load, or make a page that crawlers cannot access suitable to represent another URL. During a site move, map old URLs to relevant new destinations and test the mappings; Google’s migration guidance recommends testing redirects and updating canonical annotations and sitemaps.

In Google Search Console, URL Inspection distinguishes the site-declared canonical from the Google-selected canonical in the indexed report. The live test checks current accessibility but does not predict which URL Google will select; it also follows redirects without necessarily showing that it did so or identifying the final URL tested. Use the URL Inspection tool to compare Google’s indexed information with the page’s current response. Even after a live test or indexing request, Google might not index the URL or use its declared canonical.

Fix mismatched signals and bad destinations

A canonical is a preference, not a command. Google can disregard it when the destination is unrelated, inaccessible, or inconsistent with other signals. If one page points to one canonical while the sitemap and internal links favor another URL, correct the disagreement rather than adding more declarations. For a closer look at these problems, see Google’s canonicalization troubleshooting guidance.

A redirect should send visitors to a page that genuinely replaces the old one. Sending many unrelated retired pages to the homepage can confuse visitors, and Google may treat those redirects as soft 404s. When no relevant replacement exists, return 404 or 410. Avoid redirect chains by pointing each old URL directly to its final destination. Google says its crawlers generally follow up to 10 redirect hops for general web content, but that limit is not a target; direct redirects are easier to check and avoid unnecessary hops. A redirect loop prevents the browser from reaching a final page.

Do not use robots.txt or noindex as a substitute for canonicalization. Google advises against using robots.txt to select a canonical and does not recommend noindex for choosing a preferred URL within a duplicate set. These controls address different questions: whether a crawler can fetch a URL and whether a page should be indexed.

What the choice means for AI search

Consistent URLs help search and AI systems identify the page a publisher wants them to represent. They do not guarantee that the page will appear in an AI answer or receive a citation. In a December 2025 post, Bing’s webmaster team connected duplicate-URL handling and canonical signals with surfacing the intended page in search and AI experiences. The post provides guidance, not a measured experiment or an effect size showing how a canonical or redirect changes AI visibility. Treat URL consolidation as a way to reduce ambiguity about the page, not as a promise of AI citations.

What I'm building

Delegate tasks. Get software.

Give Vroni a GitHub issue, bug report, spec, or rough idea. It reads the repo, plans the change, writes code, runs checks, and works toward a review-ready pull request.

Take a look at vroni.com