When Not to Use Canonical Tags

Google defines a canonical URL as “the URL of the best representative page from a group of duplicate pages.” The canonicalization process selects a preferred version of content, and uses multiple signals to do so, not just rel="canonical" tags.

Canonical Tags Are Signals, Not Orders

The rel="canonical" link element, introduced by Google in 2009, gives webmasters a way to specify a preferred URL for duplicate content. However, Google treats it as a hint rather than a directive.

The tag is placed in the head section of the duplicate pages, pointing to a canonical version. But Google may still choose a different canonical based on other factors. According to Google’s canonicalization documentation, the canonical URL can come from a set of duplicates, not just the pages pointing to a rel="canonical" link.

For example, a blog post available on desktop and mobile might have a shared canonical: `` <link rel="canonical" href="https://example.com/blog-post" /> `` But if the mobile and desktop versions each have many more internal links, Google might treat the more linked version as the best representative. In a similar way, duplicate pages without a canonical may still be chosen as canonical by Google.

How Google Decides Which URL Wins

Google considers multiple signals in determining a best representative page:

1. Rel="canonical" annotations
2. Redirects
3. Sitemap inclusion
4. Internal links
5. Crawl history

If these signals point to different pages, such as a redirect leading away from a canonical, Google chooses based on a holistic assessment of the factors. A canonical that conflicts with internal links and redirects is more likely to be overridden.

Where Cross-Domain Canonicals Are Legitimate

Cross-domain canonicals are not outright forbidden, but their use is narrow. Google documents using cross-domain canonicals for duplicated content appeared across domains.

For example, a syndicated article may be published on both a blog and a news site. The rel="canonical" declaration can specify the article's origin: `` https://blog.example.com/article-title ``

If the same article is available on the syndicator and the user's site, a canonical on the syndicator's page can point to the user's. `` http://news-site.example.org/duplicate-article ``

But this is a limited use case. Unauthorized cross-domain duplication, such as scraped sites, is subject to content removal under Google's duplicate-content policies.

What Search Console Shows When Canonicals Fail

Google Search Console distinguishes the user-declared canonical from the Google-selected canonical URL. If these conflict, the tool labels the URL's status as "Duplicate, user declared canonical different."

Additionally, if Google does not choose a canonical for a set of pages, some duplicate pages may appear indexing statuses like "Duplicate, user declared canonical different." both "Duplicate, user declared canonical different" and "Duplicate, no canonical user selected."

Other indicators of issues are when a canonical page has no internal or sitemap links, or has a redirected canonical. Google Search Central also explicitly calls out redirecting the canonical as a failure case.

Common Mistakes with Canonical Tags

Despite their widespread use, canonical tags have some common pitfalls. Here are a few scenarios where a canonical tag:

1. Can Erase the Wrong Page
2. Pointing to a redirect detects the designated URL as canonical, but removes the redirected URL from search results. Google considers both the canonical and the redirect. If the redirect, not the target, is the ideal URL, such a canonical would conflict with the redirect.

1. Taking Priority Over a Redirect can cause issues.

If the canonical is placed on a redirected URL, and the redirect lands elsewhere, Google evaluates the redirect to determine the best representative page. If the target of the redirect matches a canonical signal, it is not the page declared as canonical.

1. Discarding Related Pages
2. An eCommerce site may generate product pages for each filtering and paginated option, rendering the original product page a duplicate in search results. Similar issues happen with a filtered page version generated for different sorts or filters. Pointing dozens or hundreds of filtered or paginated user pages to the main variant of a URL may lead to only that URL being indexed., a canonical that marks one page as definitive may hide the others

Other checks are:

1. Paginated series canonicals: pointing all paginated pages to page one may lead to Google indexing only the first page. Pointing all paginated pages to a canonical with no first page view.

1. Filtered or sorted pages: a canonical pointing to the most common sorting or filter can lead to Google indexing only the filtered view. -TO VERIFY]

Multiple pages may be missing from search after a canonical change; a canonical error can deindex pages.

Plug-ins may sometimes generate fixed canonicals:

An overridden canonical may result in Search Console showing missing canonical errors.

A canonical on a redirected page may not work: A canonical URL to a page that redirects is itself redirected, and is not treated as the definitive URL for this content, and is not considered the canonical URL. The canonical tag matters in tandem with other signals: Google uses the user-specified canonical and an internally determined canonical to determine the best page for a query. If the rel="canonical" and the rest of Google’s signals are inconsistent, Google determines the best URL for a query based on a combination of all signals, including, not limited to, the canonical URL.

Accordingly, canonical URLs should be placed on URLs where the content has changed minimally, and is indistinguishable in search value. Multiple distinct pages should generally avoid canonical links to a single page, or Google may deindex pages; other tools like hreflang or JSON-LD are better suited for single-page structures.