Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

SEO › Troubleshooting

Duplicate Without User-Selected Canonical

What the Duplicate without user-selected canonical status means and how to fix it: missing canonicals, parameter and protocol variants, and thin copies.
Troubleshootingintermediate6 min readUpdated

First, find out which URL Google picked

Take a few examples from the "Duplicate without user-selected canonical" row of the page indexing report in Search Console (GSC) and run each through URL inspection. Two fields matter: "User-declared canonical", which will read none or be missing, and "Google-selected canonical", which names the URL Google chose instead. Write both down side by side for every example.

Now judge the pairing. If Google picked the URL you would have picked, the status is close to harmless: the duplicate is being consolidated correctly and the task is tidying the signals so the choice is stable. If Google picked an address you never wanted shown, such as a tracking variant, a print view or an old host name, the signals are pointing the wrong way and need correcting. And if the two pages are not really duplicates at all, for instance two branch pages that differ only in a town name, the content is too similar for Google to treat them separately, which is a content question rather than a tagging one.

Diagnosis table

Diagnosis

SymptomCheckLikely causeFix
Inspected pages show no user-declared canonical at allView source and rendered HTML for a canonical link element or headerNo canonical tag on pages that exist in several versionsAdd a self-referencing canonical to the preferred version of every template
Google's chosen URL differs from the excluded one only by a query stringCompare the two addresses; list the parameters in useParameter, tracking or sort URLs creating copiesCanonicalise to the clean URL and stop linking parameter versions internally
The pair differ by http and https, www, case or a trailing slashRequest each variant and read the status code it returnsProtocol, host or trailing-slash variants not redirectedPermanently redirect every variant to the single preferred form
Two distinct pages (towns, sizes, colours) share almost all their textDiff the body copy of the pairNear-identical templated pages (locations, variants)Merge into one page, or make each version substantially different
A canonical exists, yet links and the sitemap use another addressCrawl inlinks by URL; sitemap entries; hreflang targetsCanonical signals conflicting: sitemap, internal links and tag disagreeAlign links, sitemap and annotations with the declared canonical

Fix, verify and prevent each cause

No canonical declared

Google's documentation on consolidating duplicate URLs lists the methods a site can use to state a preference and notes that, without one, Google chooses for itself. A Canonical Tag on each template is the simplest statement.

Fix: add a self-referencing canonical on every indexable page, generated from the clean, absolute URL rather than from whatever address the visitor requested.

Verify: URL inspection on the variant now shows a user-declared canonical, and on the preferred page Google's choice matches yours.

Prevent: make the canonical part of the base template so new page types inherit it, and keep the rules for canonicals and directives from crawling and indexing in the developer handbook.

Query strings producing copies

Campaign tags, session identifiers, sort orders and filter selections can each attach to a page and create an address Google may crawl. When the content is unchanged, Google clusters them.

Fix: make the canonical ignore non-content parameters, remove parameters from internal links, and keep tracking tags out of links between your own pages. Where filters create real but low-value variants, decide their index status deliberately.

Verify: the share of parameter URLs among this status's examples falls, and inspection shows Google choosing the clean URL.

Prevent: keep a register of permitted parameters and their treatment, as recommended in faceted navigation and index control. The tutorial on auditing URL parameters builds that register, one class of parameter at a time.

Host and protocol variants still answering

If the same page loads over http and https, with and without www, or with and without a trailing slash, every variant is a candidate. Google's redirects documentation describes permanent server-side redirects as a strong signal that the target should be canonical.

Fix: configure the server or CDN to redirect each variant to the single preferred form in one hop, and make the canonical tag use that same form.

Verify: request each variant and confirm it returns a permanent redirect straight to the preferred URL; inspection then shows the variant as redirected rather than duplicate.

Prevent: include host, protocol and slash rules in launch checks, the same ones the site migrations guide applies to a move.

Templated pages too similar to stand apart

Location pages that swap only the town name, or product variants that each get a page with identical descriptions, look like one document to Google. Tagging will not make them distinct.

Fix: decide whether each version deserves its own page. If it does, give it content specific to it: local details, staff, reviews, or the attributes that genuinely differ. If it does not, merge the versions into one page and redirect the rest.

Verify: after recrawl, each retained page is indexed under its own URL.

Prevent: apply the keep, merge or differentiate decision from keyword cannibalisation before generating pages from a template.

Signals that contradict each other

A canonical tag is a strong hint, not a command. If the navigation links to one version, the sitemap lists a second and the tag names a third, Google weighs them and may pick a URL you did not intend.

Fix: choose the canonical, then update menus, breadcrumbs, in-content links, sitemaps and any hreflang annotations to use exactly that address.

Verify: a crawl finds no internal links to non-canonical versions, and the sitemap contains only canonical URLs that return a success status.

Prevent: generate links and sitemaps from a single URL helper, and audit link targets on a schedule as the internal linking guide describes.

Prevention: one URL per piece of content

Duplication creeps in when URLs are generated in more than one place. Agree the preferred form of every URL (protocol, host, case, trailing slash) and enforce it at the edge with redirects. Build canonical tags, internal links and sitemap entries from the same function so they cannot drift. Strip tracking parameters from internal navigation. Decide before launch how new filters, variants and location pages will be addressed. Review this report together with the "Duplicate, Google chose different canonical than user" row, because a declared canonical that Google overrides is the same family of problem one stage later. Fewer duplicate URLs also means Google spends less of its Crawl Budget on copies. Where duplicate pages also show thin content, the sibling page on crawled, currently not indexed covers the quality side, and pages that redirect or collapse into empty templates may surface as soft not-found errors instead.

Illustrative example — not a real result

A dental group with four versions of every page

A dental group with 12 clinics finds 1,900 URLs in this status. Inspection shows two patterns. First, every page answers on four hosts and slash combinations, and Google has picked the www-without-slash form for some pages and the bare-domain form for others. Second, the 12 clinic pages share 95% of their text. The team redirects all variants to https with www and a trailing slash in a single hop, updates the canonical helper to match, and rewrites each clinic page with its own team, opening hours and directions. Within two months the count falls to about 150, mostly old campaign URLs still linked from emails.

Deciding whether a given duplicate should carry a canonical or be redirected is the subject of the comparison of canonical tags and 301 redirects.

Common mistakes

Adding noindex to duplicates, which discards their signals instead of consolidating them. Pointing every variant's canonical at the home page. Canonicalising to a URL that redirects or returns an error. Expecting a tag alone to override a navigation that links to the wrong version on every page.

Frequently asked questions

Is this status a problem if Google picked the right URL?

Not much. If Google's chosen canonical is the URL you would have chosen, the duplicate is being consolidated correctly. Tidy the signals anyway, by declaring the canonical and aligning links, so the choice stays stable and does not flip later.

Should I noindex the duplicate URLs?

Usually not. Noindex removes a URL without passing its signals to the preferred version. A redirect or a canonical tag consolidates the duplicates onto one URL instead, which is what you want when the content is the same.

Why did Google ignore the canonical tag I added?

A canonical tag is a strong hint rather than a directive. If internal links, the sitemap or redirects point at a different version, Google may weigh those signals more heavily. Make every signal name the same URL and the tag is far more likely to be respected.

Terms used on this page

Where this page is used

Also related

Sources

  1. Consolidate duplicate URLs — Google Search Central
  2. Redirects and Google Search — Google Search Central

Related