Duplicate without user-selected canonical: What It Means and How to Fix It

"Duplicate without user-selected canonical" means Google found two or more near-identical pages, saw no canonical tag saying which one you prefer, and picked one itself. Whether to act depends on its pick. If Google kept the version you want, you can leave it. If it kept the wrong one, tell Google with a canonical tag or a redirect.
What does "Duplicate without user-selected canonical" mean?
Read the status in three parts. Duplicate: Google thinks this page is a copy of another. Without user-selected canonical: no canonical tag, the line of code naming the main version, says which to keep. So Google chose. Google's help adds: "This is not an error, but is working as intended, because Google does not serve duplicate pages."
Google's help also gives the two ways out. If you think Google chose the wrong address, mark the canonical yourself. If you think this page is not really a duplicate, make its content clearly different from the page Google chose.
Do you need to fix it?
Only if Google kept the wrong page, or if the two pages should not count as copies at all. On a small site, a duplicate can simply be the same page reached by more than one address, and then Google keeping one of them is what you want. The table helps you decide which case you are in.
What you find | Leave it or act | Why |
Google kept the address you would have chosen | Leave it, or add a canonical tag to make it explicit | Your preferred page is the one in search |
Google kept an http, non-www or parameter version | Act: redirect to your main address, or add a canonical tag | The address people see in results is not your main one |
Two different pages Google treats as one | Act: make the content clearly different | Google says the content should differ substantially |
An old copy you no longer need | Act: redirect it to the main page | Google suggests redirects for duplicates you want gone |
A tag or category page that repeats posts | Leave it, or keep it out of search | It was not meant to rank |

From Capron Media: the free Lite subscription

Is your site in good shape? See it in plain words
See what on your site needs fixing, counted by urgency, with the top three in plain words. Lite does not include the work on your site. Starter does.
How to see which page Google chose
URL Inspection shows it. Inspect the address listed in the report, open Page indexing, and read two lines: "User-declared canonical", which will be empty for this status, and "Google-selected canonical", which shows the page Google kept instead. Then open that page in your browser and decide whether it is the one you want people to land on.
In Search Console, open Indexing, then Pages, and click "Duplicate without user-selected canonical".
Click the magnifying glass beside an example address.
Under Page indexing, read "Google-selected canonical".
Open both addresses side by side and compare what a visitor sees.
Repeat for a few examples. If Google keeps choosing the same kind of address, such as the http version, the cause is site-wide.
One caution from Google: duplicate and canonical conditions are not tested in the live test, so read the indexed result, not the live one.
The three duplicate statuses, side by side
Search Console has three statuses about copies of a page, and the difference is who chose the main version. In this one, nobody told Google, so it chose. In the other two, you told Google: once it agreed with you, once it did not. The table puts them side by side.
Status | Did you name a main version? | What Google did | What to do |
Alternate page with proper canonical tag | Yes | Followed your tag | Leave it |
Duplicate without user-selected canonical (this page) | No | Chose a main version itself | Check its choice |
Duplicate, Google chose different canonical than user | Yes | Chose a different main version | Check your other signals |

Where duplicates come from on small sites
The same page can be reachable at several addresses without you ever making a copy. Google's own list of reasons for duplicates includes the HTTP and HTTPS versions of a site, sorting and filtering options, and a demo or test version left open by accident. On a small site, check these first, using the table below.
Source | Example addresses | How to fix |
http and https | http://example.com/menu and https://example.com/menu | Redirect http to https |
www and non-www | example.com/menu and www.example.com/menu | Redirect one to the other |
Trailing slash | /menu and /menu/ | Redirect one to the other |
Parameters | /menu?ref=newsletter | Canonical tag on the page, naming /menu |
Tag and category pages | /blog/tags/bread listing the same posts as /blog | Keep them out of search, or give them their own text |
A test or old copy | /menu-old or a copy on a staging address | Redirect it, or remove it |
When we checked our own site on November 2, 2026, the http, non-www and trailing-slash versions of our addresses each ended, through permanent redirects, at one https www address. That is the setup that stops the first three rows from ever reaching this report.

How to fix "Duplicate without user-selected canonical", step by step
Decide on one main address for each page, then make every signal point at it. Google calls a redirect the strongest signal and a canonical tag a strong one; listing the address in your sitemap is a weaker one. Use a redirect when the copy has no reason to exist, and a canonical tag when it does.
For each listed page, choose the main address you want in search.
If the copy has no reason to exist (an old page, an http or non-www version), redirect it to the main address with a 301.
If the copy must stay reachable (a parameter or tracking version), add a canonical tag on it naming the main address.
Give the main page a canonical tag naming itself.
Make your menus, internal links and sitemap use only the main address.
If the two pages are really different, rewrite one so its content differs clearly, and give each its own canonical.
Inspect the main address and click Request indexing.
Step 5 is the easiest one to miss. If your own links point at the copy, you are telling Google the copy matters.
Fixing it on Wix, WordPress, Squarespace and Shopify
The two tools you need are a canonical tag and a redirect, and each platform puts them in a different place. Below is where each one lives, with the names taken from each platform's help pages as they read on November 2, 2026. Where a platform sets a canonical for you, check that first.
Wix. Wix gives every page a default canonical tag pointing to itself, and its help recommends keeping it. If you do need a custom one, it is under SEO basics, then the Advanced SEO tab, then Additional Tags, in the page's settings. For redirects, Wix's URL Redirect Manager, in the SEO dashboard under Tools and settings, creates 301 redirects. Wix adds a redirect automatically when you change a page's URL slug, but not for blog posts, which need one made by hand.
WordPress. WordPress adds a canonical tag to single posts and pages by itself. Duplicates can come from elsewhere, such as tag archives or parameters. Your SEO plugin, if you use one, controls what those other pages carry; check its documentation rather than guessing.
Squarespace. Redirects go in Developer tools, then URL mappings, as one line per redirect: the old path, an arrow, the new path and 301. Squarespace's help says a 301 only works when the old address no longer exists, so delete, disable or rename the old page first.
Shopify. Redirects are under Content, then Menus, then View URL Redirects, then Create URL redirect. Shopify's help warns that you can redirect only from a broken address: "If the URL still loads a valid webpage, then the URL redirect won't work." For a duplicate that still loads, that means removing the old page first, or leaving the duplicate to a canonical tag.
How to confirm the fix (and how long it takes)
Inspect the copy again after Google has recrawled it. A redirected copy will move to "Page with redirect". A copy with a canonical tag will move to "Alternate page with proper canonical tag". Either change means Google has your instruction. Google says it "might hold pages in a duplicate cluster for up to two weeks" after a fix.
For a group of pages, click Validate fix on the row once all of them are fixed. Validation runs from Started to Passed or Failed, and Google's help says it "typically takes up to about two weeks." If you made two pages genuinely different instead, request indexing for both; Google says pages split out faster when the difference is "clear and significant".
Other "page not indexed" reasons, set against this one
Two of the other reasons are close relatives of this one and appear in the comparison above. The rest are about Google being stopped, or about a page it read and set aside for other reasons. This table says, for each, what separates it from a duplicate Google sorted out on its own.
Status | What separates it from this one |
Alternate page with proper canonical tag | You named the main version, and Google agreed |
Duplicate, Google chose different canonical than user | You named the main version, and Google picked another |
Crawled - currently not indexed | Google left the page out without calling it a copy |
Discovered - currently not indexed | Google has not read the page, so it cannot judge it a copy |
Page with redirect | The address forwards on, which is one way to settle a duplicate |
Excluded by 'noindex' tag | The page asked not to be indexed |
Blocked by robots.txt | Google was not allowed to fetch it |
Soft 404 | The page looked empty, not copied |
For the content side of duplicates, our guide to whether repeat information on a website is bad for SEO goes further.
Ways to catch duplicate pages before Google sorts them
Duplicates are easiest to fix before Google has to pick between them. A crawl of your own site lists pages with the same title or the same text, and shows which ones carry a canonical tag. These options run from Google's own reports, which show what has already happened, to checks that look ahead.
Option | What it shows about duplicates | Cost |
Page indexing report | Which addresses Google filed as duplicates, as a sample | Free |
URL Inspection | The page Google kept for one address | Free |
A crawler, such as Screaming Frog | Exact and near-duplicate pages, and each page's canonical tag, among the pages it crawls | Free version crawls up to 500 URLs, as of November 2, 2026 |
Capron Media (ours) | Free check, with a free Lite subscription: a short, plain read of a website: what to fix first across the site, on its home page, and in its local results | Free |
If you would rather have the whole site read, with every finding and its fix, our SEO audit services page describes the SEO Audit, which a person checks before it reaches you. See our pricing page for what it costs. The guides to an SEO audit and a technical SEO audit explain the rest.
Questions about duplicates without a canonical
Does a duplicate page hurt my site?
Google says some duplicate content on a site "is normal and it's not a violation" of its rules. The cost is practical: people may see the wrong address in results, and it can be harder to track how your content performs.
Should every page have a canonical tag?
A canonical tag that names the page itself is a sensible default, and the platforms add one for you: Wix on every page, WordPress on single posts and pages. Where you do set one by hand, Google asks that it match your redirects and sitemap.
Should I use noindex on the duplicate instead?
Not to choose between copies. Google advises against noindex for picking a canonical within one site, because it removes the page from search entirely. A redirect or a canonical tag tells Google which copy to keep without hiding anything.
Why does Google think two different pages are duplicates?
Google compares the main content of each page. If two pages share most of their text, such as two service pages with only the town name changed, Google can cluster them. The fix is to make each page say something the other does not.
From Capron Media: the free Lite subscription

See what to fix first on any website
Your own site doesn't use up a check. You can check up to three other sites, each once every 30 days. Lite does not include the work on your site. Starter does.
Written by Spencer Capron, founder of Capron Media, a Middleborough, Massachusetts SEO service for small business owners. Spencer is HubSpot SEO II certified, holds CompTIA Security+, and has run recurring SEO programs for 30+ small-business clients. He wrote this post in collaboration with our AI agents, and he reviews everything published here.





Comments