Duplicate without user-selected canonical

The status sits under Why pages aren’t indexed in the Page indexing report. Google’s help page says the page is a duplicate of another one and doesn’t indicate a preferred canonical, so Google chose the other page as canonical and won’t show this one in search. In its own words, it’s “not an error, but is working as intended”.

A canonical is the one address Google keeps for a piece of content. It collects the links and signals of all the copies onto it, and that’s the address people see in results. “User-selected” means you told Google which one you want. Here, you didn’t, so Google guessed.

Look at the URLs in the report. The answer depends on whether they are addresses you want in search.

Before changing anything, check what Google actually did:

• In the report, click a URL and choose Inspect URL.
• Open Page indexing. Compare User-declared canonical with Google-selected canonical.
• “User-declared: None” is this status. If Google’s choice is the address you want people to land on, you’re done.

If you did declare a canonical and Google picked a different one, the status is a different one: “Duplicate, Google chose different canonical than user”. There Google disagreed with your choice, which is worth a closer look.

• The same site at several addresses. http and https, www and without, with and without a trailing slash. Each is a separate URL to Google until a redirect says otherwise. Trakis checks http (CRAWL-009) and a second domain that mirrors the first (INDEX-006).
• Parameters. Tracking tags, sorting, filters and session IDs turn one page into dozens, and links to them inside your own site make it worse. Session IDs in the address are CRAWL-005.
• No canonical tag at all. Without one, every copy is equally a candidate and Google has to guess. That’s INDEX-003; a relative canonical (INDEX-005) is easy to break.
• Pages that are nearly the same. Product variants, tag and category pages, location pages with only the city name changed. Google can treat them as duplicates even at different URLs (CONTENT-008).
• A sitemap that lists the copies. Google reads the sitemap as a hint about which URLs are canonical; listing the duplicates muddles it (CRAWL-010).

Google’s guide to canonical URLs ranks the signals: redirects are the strongest, a rel="canonical" link is strong, and the sitemap is the weakest. Use them in that order:

• Redirect the addresses you’re retiring. A permanent (301) redirect from http to https, from www to the bare domain or the other way, and between slash and no slash.
• Put a canonical on every page, pointing at itself. Use the full address, not a relative path: <link rel="canonical" href="https://example.com/shoes/">. Parameter versions then declare the clean URL as theirs.
• List only canonical URLs in the sitemap, and link to them, not to their copies, from your own pages.
• Don’t use robots.txt or noindex to pick a canonical. Google says so plainly. A copy blocked in robots.txt can’t be read, so Google can’t see it’s a copy, and noindex takes the page out of search altogether.

Then inspect a few of the URLs again, and press Validate fix in the report if it’s offered. Google rechecks the pages as it recrawls them, so the count goes down over days or weeks, not at once.

Is it an error? No. Google says it’s working as intended. It becomes a problem only when the page left out is one you want found.

Will it hurt my rankings? Not by itself. The original still ranks, with the copies’ signals added to it. The risk is Google choosing a copy you don’t want as the one it shows.

Should I delete the duplicate pages? Only if nobody needs them. Usually a redirect or a canonical is enough, and the page keeps working for visitors.

How do I check if a URL is canonical? URL Inspection in Search Console shows both the canonical you declared and the one Google chose.

The Trakis audit checks every page it crawls for the causes above: missing and relative canonicals, http and mirror domains, session parameters, near-duplicate text, a sitemap that lists the wrong URLs. It counts each problem once, with the pages it covers. Create a free account and add your site, or read every check it runs. If robots.txt is part of the story, the robots.txt validator shows which line decides.