Skip to content

"Duplicate without user-selected canonical": what it means and how to fix it

Google sees the page as a copy of another page, and the page doesn't say which one is the main page.

You can fix this

By Bruno Amorim · ·

What it means

Google found another page with the same or almost the same content and picked that one as the main page. This page has no canonical tag, so Google chose for you.

The fix, in short: Add a rel="canonical" tag to each copy that points to the page you want on Google. Use that same address in your links and your sitemap.

Where to see it

In Search Console, open Indexing > Pages, click the Duplicate without user-selected canonical row, and inspect one example address. In URL Inspection, open Page indexing and read the last two lines.

In Search Console: URL Inspection > Page indexingExample
URL
https://example.com/shoes/trail-runner?color=blue
Page indexing
Page is not indexed: Duplicate without user-selected canonical
User-declared canonical
None
Google-selected canonical
https://example.com/shoes/trail-runner

The page names no main address, so Google picked one. Here Google picked the clean product page, which is probably what you want.

Illustration, not a screenshot. Your addresses, dates, and counts will differ.

Is it a problem? Check in one minute

  1. Is Google-selected canonical the page you want on Google?

    Google's help calls this "working as intended". This page is a copy, and Google shows the main one.

    If so: Nothing is broken. You can still add a canonical tag to the copies. URL Inspection's help recommends declaring the canonical when a page is one of a set, so Google doesn't have to guess.

  2. Did Google pick the wrong page?

    For example, Google picked the ?color=blue version, an old address, or the http version.

    If so: Tell Google your choice with a canonical tag or a 301 redirect. The table below matches the fix to the pattern.

  3. Is this page not a copy at all?

    Two pages that serve different searches, such as two city pages built from one template, can look like copies to Google when only the name changes.

    If so: Google's help says to make sure the content differs substantially between the two pages.

Find the pattern behind the copies

Export the example list and read the addresses side by side. Most copies come from a few patterns.

PatternExampleBest fix
http and httpshttp://example.com/a and https://example.com/a301 redirect every http address to https.
www and no wwwexample.com/a and www.example.com/a301 redirect to the version you use.
Slash at the end/shoes and /shoes/301 redirect to one style, and use it in your links and sitemap.
Tracking and session parameters?utm_source=news, ?sessionid=123A canonical tag pointing to the clean address.
Sort and filter addresses?sort=price, ?color=blueA canonical tag pointing to the main list. Link to the clean address.
One product in several categories/men/trail-runner and /sale/trail-runnerOne product address. A canonical tag on the others.
Print and AMP versions/guide/print, /guide/ampA canonical tag pointing to the main page.
Capital letters and index.html/About and /about, /index.html and /301 redirect to the lowercase, clean address.
Staging or demo site left openstaging.example.com/aPut the staging site behind a login.

Pick the fix

Canonical tag: when both addresses must keep working

Put the tag in the <head> of every copy, and on the main page itself, with the full address. Google's docs call rel="canonical" a strong signal. Put it in the HTML your server sends. Google's docs say not to let JavaScript change it.

In the <head> of each copy and of the main page
<link rel="canonical" href="https://example.com/shoes/trail-runner" />

301 redirect: when the copy has no reason to exist

A redirect sends people and Google straight to the main page. Google's docs call redirects a strong signal, and say server-side redirects take effect fastest.

Different content: when the pages serve different searches

Give each page its own main content: local details, prices, photos, answers to the questions people ask about that city or product. Changing the name in the title and first line leaves the main content the same.

Make every signal name the same page

Link to the main address in your menus and text, and list only that address in your sitemap. Google's docs say not to name different main pages with different methods, such as one address in the sitemap and another in the tag. Don't use robots.txt or the Removals tool to pick a main page either. Google's docs advise against both for this.

CrawlCoach treats the pages you track as pages you want on Google. When Google sets one aside as a copy, the fix list shows its live canonical tag next to the page Google chose. If the tag is missing or points elsewhere, the fix is a tag that names the page itself, and the next check confirms it. The fix list also flags sitemap addresses whose canonical tag points to another page.

See which of your pages Google treats as copies, and the 3 to fix first.

Get my free report

What Google has said

"This is not an error, but is working as intended, because Google does not serve duplicate pages."
Search Console Help, Page indexing report. Source
"Some duplicate content on a site is normal and it's not a violation of Google's spam policies."
Google Search Central, What is URL canonicalization. Source

Copies break no Google rule. The cost is that Google may show the address you didn't want, and your search data is split across addresses.

"indicating a canonical preference is a hint, not a rule."
Google Search Central, What is URL canonicalization. Source

That is why your links, sitemap, and redirects must agree with the tag.

How long it takes, and what to measure

Google's canonical troubleshooting guide says that even after you fix the content, Google might hold pages in a duplicate cluster for up to two weeks. Redirects and tags count once Google crawls the copies again.

Test live URL can't check this reason. Google's help says canonical selection isn't tested in the live test. Check your HTML yourself, then watch these:

  • The page count on the Duplicate without user-selected canonical row. It should fall.
  • Where the copies go. Copies you redirected move to Page with redirect. Copies with a canonical tag to the main page usually move to Alternate page with proper canonical tag. Both mean the fix worked.
  • Google-selected canonical in URL Inspection, for a few of the copies.

If "Validate fix" failed: Validation fails here when Google still finds copies with no canonical tag. Check that the tag is in the HTML your server sends, and that every copy has it, not only the main page. See what each validation result means.

Fixes by platform

On WordPress

WordPress adds a canonical tag to single posts and pages on its own. Archives, page 2 of lists, and attachment pages are where copies pile up. SEO plugins such as Yoast SEO and Rank Math add canonical tags to more page types. Rank Math also has a Canonical URL field in each post's Advanced tab. More in WordPress site not showing on Google.

On Next.js

Next.js adds no canonical tag until you set one. Set metadataBase in your root layout, then set alternates: { canonical: ... } in each page's metadata or generateMetadata. Keep the address the same as the one trailingSlash produces, so the tag doesn't point at a redirect. More in Next.js pages not indexed.

Also common on

This reason is one of the most common on Hostinger, v0, and Replit. Each guide shows the cause and the fix in that platform's own settings.

Questions

Is "Duplicate without user-selected canonical" bad?

Not always. If Google picked the page you want as the main one, this page is just a copy. Inspect it in URL Inspection and look at Google-selected canonical. Act only if Google picked the wrong page.

Why does a page that redirects show this reason?

Google saw the page before the redirect existed, or the redirect doesn't run for every visitor. Open the address in a private window to check that it redirects now. The report changes after Google visits the address again.

Should I add noindex to the copies instead?

No. Google does not recommend noindex for picking a main page within one site, because it removes the page from search completely. Use a canonical tag or a 301 redirect.

Related guides

Sources

Checked on October 8, 2026.

Which of your pages have this problem?

Get my free report

Free for up to 500 pages. No card needed. Sign in with the Google account you use for Search Console.

Related problems