Duplicate, Google chose different canonical than user
You named this page as the canonical of a group of similar pages, and Google picked another one of them to index instead. Sometimes Google’s choice is the better one. When it is not, the cause is usually the site saying two things at once.
What Google is telling you
Google’s help page says the page is marked as canonical for a set of pages, but Google thinks another URL makes a better canonical, and indexes that one. To see which, inspect the URL: under Page indexing, compare User-declared canonical with Google-selected canonical.
Read those two lines from the inspection of the indexed page. Google says the live test does not check duplicate or canonical conditions, so a live test that looks fine says nothing about this status.
First: is Google’s choice wrong?
Google’s canonicalization guide asks you to check whether the URL it chose makes more sense for someone arriving from Search than the one you preferred. It may: the https address rather than the http one, the address without parameters, the version your own menu links to. Then the fix is to agree with it: make that address the one your canonical, your links and your sitemap all name.
When Google picked an address you really do not want shown, look for the cause below.
Why Google picks another address
| The cause | What to do |
|---|---|
| The site says two things. The canonical names one address while the sitemap lists another, your links go to a third, or a redirect leads elsewhere. | Make every signal name the same address. Google says the signals stack, and asks you not to name different canonicals for one page by different methods. |
| The pages are near-duplicates: product variants, the same text under two addresses, a print version. | Make them clearly different if they are meant to be separate pages, or accept one canonical for the group. |
| The canonical names a page that is not really the same. | Google says it will never choose a declared canonical that is not similar to the page. Point it at the true equivalent, or at the page itself. |
| http and https both answer. | Google prefers https as canonical. Redirect http to https. |
| Language or regional versions with the same text. | Add hreflang, and give each version a canonical in its own language. |
| A plugin or theme writes the canonical or a redirect wrongly. | Look at the page’s HTML in the browser’s developer tools, and report it to whoever makes the plugin. |
When the address Google chose is on someone else’s site, the guide names three more causes: a hacked page that redirects or names a canonical elsewhere, content syndicated to partners (for which it suggests the partners block indexing of their copy), and a site copying yours.
What the Trakis audit finds
- Canonicals that name another page, with the address each one names (INDEX-004): the list to read against Google’s choices.
- Pages that repeat each other’s text, the near-duplicates Google groups together (CONTENT-008).
- A sitemap that lists addresses it should not, among them pages that redirect or name another canonical (CRAWL-010).
- Internal links written to another address of the same page: the other host, with or without www, or http (LINK-006).
- http serving the site instead of redirecting to https (CRAWL-009).
What not to do
Google’s guide to specifying a canonical rules out four shortcuts:
- robots.txt. Blocking the address Google chose does not make it pick yours; it only stops Google reading it.
- The URL removal tool. It is not a canonical setting.
- noindex on the page Google picked, to steer the choice within your own site. Google does not recommend it.
- Different canonicals by different methods, such as one in the HTML and another in an HTTP header.
After the fix
Inspect the most important pages again and use Request indexing; Google notes it has a quota, so keep it for the pages that matter. Google says pages can stay grouped as duplicates for up to two weeks after the content is fixed, and split sooner when the differences are clear.
Common questions
Is my page out of Google? This address is. Google shows the one it chose in its place, with the group’s signals added to it.
Why does Google ignore my canonical? A canonical is a strong signal, not an order. When the other signals point elsewhere, or the pages are not as alike as the tag says, Google weighs them all.
How is this different from the other two canonical statuses? In Alternate page with proper canonical tag, Google agreed with your canonical. In Duplicate without user-selected canonical, you declared none and Google chose. Here you declared one, and Google chose another.
Check the whole site
The Trakis audit reads the canonical of every page it crawls, with the sitemap and the internal links beside it, and reports where they disagree, each problem once with the pages it covers. Create a free account and add your site, or see every Search Console status, explained.
Last changed 28 October 2026.