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

Google’s Page indexing report describes the status this way: the page is a duplicate of another page, and it does not indicate a preferred canonical. Google adds that this is not an error, because it does not serve duplicate pages. One version is indexed, and the rest are listed here.

Duplicates are rarely written on purpose. They are produced by how a site is built.

• The same page answering at several addresses: with and without www, with and without a final slash, in capitals and lower case.
• Addresses with tracking or sorting added to them.
• Product pages for the same item in several categories.
• Printer-friendly and mobile versions.
• Pages that were meant to differ and carry the same text, such as one page per town with only the town’s name changed.

If the addresses listed are technical variants and the version Google kept is the one you would have chosen, nothing needs doing. The outcome is the one a canonical tag would have produced.

Three cases deserve a fix.

• Google kept the wrong version. Inspect the address: URL Inspection shows which page Google selected as canonical.
• The duplicates are real pages that should each rank. They need content of their own.
• The list is long and growing, which means the site is producing addresses without limit.

Tell Google which version is the original, in one of three ways.

1. Add a canonical tag to every version, naming the preferred address.
2. Redirect the duplicate addresses to the preferred one. This is the stronger signal, and the right one for www and slash variants.
3. Link internally only to the preferred address, and list only that address in the sitemap.

For pages that were meant to be different, rewrite them until they are. The related status where a page did name an original is covered in alternate page with proper canonical tag, and the case where Google overrules your choice in Google chose a different canonical.