Duplicate without user-selected canonical: Google folded a copy you never named

Read the status as a missing preference

‘Duplicate without user-selected canonical’ means Google fetched a URL, judged it to be a copy of another page, and found no canonical tag it could treat as your choice. The report shows a Google-selected canonical and no user-selected one. The duplicate stays out of the index. The selected URL is the one that can rank.

This sits between two statuses teams already know. ‘Alternate page with proper canonical tag’ means you named a canonical and Google agreed. ‘Duplicate, Google chose different canonical than user’ means you named one and Google picked a different URL. Here nobody on the site made the choice, so search did. It is also a GEO issue: an answer engine cites the URL Google kept. A passage left on an unnamed duplicate does not travel, even when that copy is the page a person would have sent to a colleague.

• Confirm the report shows no user-selected canonical.
• Keep this status separate from a canonical Google refused.
• Treat the Google-selected URL as the one search, and often an answer engine, will reuse.

Find the copies that never declared a winner

Open the sample URLs and write the Google-selected canonical beside each one. The usual pattern is a template with no rel=canonical at all, or a tag Google ignored because it was empty, relative, or pointed at a URL that redirects. Host pairs, trailing-slash pairs, print views, and parameter copies land here when the preferred address was never named. On a Hong Kong site, an English page and a Traditional Chinese page that repeat the same sentences, and declare nothing, can be folded the same way.

Compare the bodies before you add a tag. If the two URLs are the same document, the status is describing a real duplicate and the missing preference is the bug. If they answer different questions — a district service page against the citywide one, or a zh-HK guide with different examples — they should not share a canonical. A tag added only to clear the row will file a page you still need as an alternate, or start a disagreement when Google refuses the merge.

Name the URL you want kept, then match the other hints

Pick the address you would send to a colleague. Give it a self-referencing canonical. Point the true copies at that URL. Internal links and the XML sitemap should name the winner, and exact copies that do not need their own address should redirect to it. Leave the language version that answers a different query on its own canonical, with a first paragraph that could not be swapped. Decide the host, the trailing slash, and whether English and Traditional Chinese are separate documents before you repeat the tag across the template.

Request indexing only for the URL you declared, after the HTML Google will fetch is the HTML you want cited. Put the bounded answer on that URL, with the organisation and the place next to the claim, so the passage an answer engine lifts is the one search will keep. Watch URL inspection until the duplicate row clears on the copies and the winner reports your canonical as the selected one. Another request on a copy that never named a preference only asks Google to index a URL it has already folded.

• Declare one canonical, then align links, sitemap, and redirects with it.
• Keep pages that answer different queries on their own URLs.
• Request indexing for the URL you named, not for every row in the report.

What does Duplicate without user-selected canonical mean?

Google found that the URL duplicates another page, and the site did not declare a canonical Google could use. Search selected a representative and left the duplicate out of the index.

How is this different from Google chose different canonical than user?

In that status you declared a canonical and Google indexed a different URL. Here there is no user-selected canonical at all. One is a disagreement. The other is a missing preference.