Canonical Tag vs 301 Redirect: 3 Fast Ways to Choose
You have two URLs showing the same page and you need to pick one. The choice between a canonical tag, a 301 redirect and noindex comes down to one question: should the duplicate keep working for visitors? If yes, use the canonical link. If no, redirect. If the page should not appear in search at all, use noindex.
Each option sends a different signal, and a canonical tag mixed badly with the others cancels your own work. Let us break down all three, then match them to the situations you will actually meet.
What each option, including the canonical tag, tells search engines
Google describes these as signals that express your preference. None is a command, and the search engine decides which page is the duplicate. Still, the signals differ in strength and in what visitors experience.
| Trait | Canonical link | 301 redirect | Noindex |
|---|---|---|---|
| Visitor reaches the duplicate URL | Yes, page loads normally | No, sent to the target | Yes, page loads normally |
| Signal to search engines | Strong hint to index the target | Strong signal that the target is canonical | Remove this page from results |
| Where it lives | link element in the head | Server response, status 301 | Meta robots tag or X Robots Tag header |
| Passes ranking signals to another URL | Intended to consolidate them | Yes, consolidates to the target | No, it drops the page |
| Best for | Near duplicates that must stay live | Moved or retired pages | Pages with no search value |
Google also lists sitemap inclusion as a weak signal, useful but not enough on its own. Combining a redirect or canonical tag with a clean sitemap strengthens the message.
How a canonical tag works
A canonical tag is a single line in the head of a page. It names the URL you want treated as the main version. The duplicate stays reachable. Visitors never notice.
A canonical tag for a tracking URL
<!-- on https://shop.example/shoes?utm_source=news -->
<link rel="canonical" href="https://shop.example/shoes">Use a full address with the protocol and host. Google supports relative paths but says they can cause trouble later, for example when a test site gets crawled. Also keep to one canonical link per page. If a page declares two, search engines may ignore both.
Putting a canonical link on every page that points to itself is standard practice. It acts as a safety net when a campaign or a tool later adds parameters to the address.
How a 301 redirect works by comparison
A 301 is a server instruction: this page has moved for good, go here. The old address stops serving content. Browsers, crawlers and links all end up at the target. Google treats a redirect as a strong signal that the target is the canonical page.
That makes it the right tool whenever the duplicate has no reason to exist. Typical cases are switching from http to https, picking www or non www, and replacing an old product URL after a migration. Redirect each old URL to its closest equivalent. Sending everything to the homepage looks lazy to users and to crawlers.
How noindex works
Noindex is a different kind of instruction. It does not choose a preferred version. It says keep this page out of results. Google advises against using it to pick a canonical within one site, because it removes the page from search completely.
It fits pages that have a job for visitors but no value in search: internal search result pages, thank you pages, filtered views you do not want listed, and login screens. Do not block those pages in robots.txt as well, because a crawler that cannot fetch the page never sees the noindex.
Which fits your situation
| Scenario | Best choice | Why |
|---|---|---|
| http version of a page that also exists on https | 301 redirect | The insecure URL has no reason to live |
| www and non www both resolve | 301 redirect | Pick one host and send the other to it |
| Product page reached with tracking parameters | Canonical link | Parameters must keep working for campaigns |
| Print friendly copy of an article | Canonical link | Visitors may still need the print version |
| Page two of a category list | Self referencing canonical link | Each page has its own content, so do not point it at page one |
| Article republished on a partner site | Cross domain canonical tag to your original | The partner copy stays live and credits the source |
| Retired product with a close replacement | 301 redirect | Moves visitors and signals together |
| Internal search results page | Noindex | Useful to users, worthless in results |
| Staging copy of the site | Noindex plus a login | Keep it out of results and out of reach |
Who should pick which
Your duplicates are usually tracking parameters, category and tag archives, and print or amp style variants. A self referencing canonical link on every post covers most of it. Add a 301 only when you change a post slug.
If you republish posts on other platforms, ask for a canonical link pointing back to your original. Without it, the other site can outrank you for your own words.
Filters and sort orders create thousands of near duplicate URLs. A canonical hint from each filtered view to its parent listing keeps them usable for shoppers. Keep the tag on product variants too, pointing to the main product.
When a product is discontinued, use a 301 to the closest replacement. If there is no real match, a plain 404 or 410 is more honest than a redirect to the homepage.
Redirects do the heavy lifting. Map each old URL to its new equivalent with a 301, then make sure the new pages carry a canonical link that points to the new address, not the old domain.
A common slip is a template that still prints the old domain in the canonical link. Scan a sample of new pages straight after launch.
Mistakes that break a canonical tag
Checking that a tag exists is easy and tells you little. The costly failures are the ones you cannot see from a single page. The checker follows the target and looks for these conditions.
| Problem | What goes wrong | Fix |
|---|---|---|
| More than one canonical link | Search engines may ignore all of them | Keep one, remove the extra from plugin or theme |
| Canonical points to a redirecting URL | Google may not follow the chain reliably | Point straight at the final address |
| Chain from A to B to C | Signals get lost along the way | Point A directly at C |
| Protocol mismatch, http against https | Canonical favours the insecure version | Use https in the tag |
| www mismatch | Ranking signals split across two hosts | Match the host you redirect to |
| Noindex plus canonical elsewhere | Contradictory orders, noindex usually wins | Choose one instruction |
One says remove this page. The other says merge its signals into another page. Google generally honours the noindex and the canonical link then does nothing. Pick the instruction you actually mean.
You cannot redirect a page on someone else domain, and noindex on your original would remove the version you want to rank. A cross domain canonical tag on their copy credits your original while their page stays live.
A quick routine to confirm your canonical tag
The checker fetches pages through a public proxy, which may not see exactly what Google saw. Treat it as a fast first pass. Confirm anything important in Search Console. For related housekeeping, review the guide to robots.txt, allow, disallow and sitemap rules, build a clean file with the XML Sitemap Generator, and see how to tag campaign links without creating a mess of parameter URLs.
Paste a page address to see its canonical tag, whether it points to itself or elsewhere, and any redirect or noindex conflict.
Open the Canonical Tag CheckerFrequently asked questions
What is the difference between a canonical signal and a 301 redirect?
A canonical link keeps both URLs live and tells search engines which one to index. A 301 redirect sends visitors and crawlers to a new URL, so the old one stops loading its own content.
Can Google ignore my canonical tag?
Yes. Google treats it as a strong signal, not a command. If internal links, redirects or other signals point strongly elsewhere, it can choose a different canonical.
Should every page have a canonical tag?
Yes, a self referencing one is standard practice. It guards against accidental duplicates created by parameters, session IDs or testing tools.
Can I use a canonical tag and noindex together?
You should not. The two instructions contradict each other, and the noindex generally wins, which makes the canonical link useless.
Should paginated pages point their canonical tag at page one?
No. Each page in a series has different content, so give each one a self referencing canonical link. Pointing all of them at page one hides the items on later pages.
Does a canonical tag work across different domains?
Yes. A cross domain canonical hint is the usual way to credit an original when the same article appears on a partner or syndication site.
Sources: Google Search Central: How to specify a canonical URL with rel canonical and other methods