“Duplicate without user-selected canonical” in Google Search Console

What the status means

Google found several URLs with the same or nearly the same content. This one declares no canonical, so Google picked another URL from the group as the canonical and will show that one in search instead. Google's help calls it “not an error”: Google doesn't serve duplicate pages.

The key word is without user-selected: you didn't tell Google which version you prefer, so it chose for you.

Is it a problem?

To see what Google chose, open URL Inspection for one of the URLs: the Google-selected canonical field shows the URL Google indexes instead.

Why it happens

1. No canonical tags at all. Common on custom and older sites, and on headless sites where nobody wired canonical metadata.
2. The same page at several URLs: http and https, with and without www, with and without a trailing slash, upper- and lowercase paths, index.html.
3. Parameters: tracking (utm_, gclid), sorting, filtering, session IDs.
4. Pages that are too similar. City or service pages built from one template with a few words changed, product variants with identical descriptions. Google treats them as duplicates even though they are separate pages to you.

How to fix it

For true duplicates (the same page at several URLs):

• Add a self-referencing canonical to every indexable page: its own full URL. Then every duplicate URL inherits the canonical of the page it renders.
• 301-redirect the variants nobody needs: http to https, www to non-www (or the reverse), missing slash to slash. Google calls a permanent redirect a strong signal for the target to become canonical.
• Link internally only to the canonical URLs, and list only canonical URLs in the sitemap. Google calls sitemap inclusion a weak signal, but a sitemap that disagrees with your canonicals sends mixed messages.
• Keep tracking parameters out of internal links. Use them only on external campaign links.

For pages that are separate but too similar:

• Make them different in substance, not in a few words: local details, prices, examples, reviews, FAQs that differ for each page.
• Or merge them. If two pages really answer the same need, keep one, 301-redirect the other and move the useful content over.

Check your setup

The checker below shows each URL's redirect chain and canonical. Try the variants of one page, http://, www., without the trailing slash, with a ?utm_source=test parameter, and check that:

• the variants redirect to one URL, or
• they load but name the same canonical.

If each variant loads with no canonical, that is the cause of this status.

What not to do

• Don't noindex duplicate URLs to pick a canonical. Google's documentation says to use rel="canonical" for that; noindex removes the page from search entirely.
• Don't block parameter URLs in robots.txt as a canonical fix. Google can't crawl them to see they are duplicates, and blocked URLs can still be indexed without content.
• Don't canonicalize different pages to one URL just to clear the report. If they are different pages, make them different; if they aren't, merge them.