Canonical tag: real cases and errors on a SaaS site

What is a canonical URL for Google?

It is the URL Google chooses as most representative of a group of pages with identical or very similar main content. Google's documentation on canonicalization describes the process: during indexing, Google groups pages whose main content is highly similar, then selects as canonical the one it judges most complete and useful. The canonical is crawled more regularly, duplicates less often.

Two details change how the topic should be approached. First, duplicate content on a site is normal and does not violate Google's spam policies. Second, Google says it is generally not critical to specify a preference: the tag helps choose the URL shown in results, consolidate signals from duplicates and simplify tracking, not avoid a penalty.

Tag, redirect or sitemap: which signal weighs most?

Google ranks the methods itself in its page on consolidating duplicate URLs. Redirects and rel="canonical" are strong signals, sitemap inclusion a weak one. They stack: the more they point to the same URL, the more easily Google understands your preference.

Consistency is the point to watch. Google asks you not to specify different canonical URLs for the same page through different methods, for example one URL in the sitemap and another in the tag. That is exactly what happens when a plugin-generated sitemap lists URLs with a trailing slash while the template declares the version without one.

When does a SaaS site actually need one?

When the same page stays reachable through several URLs and has to remain so. Google lists five sources of duplicates; on a SaaS site they take very recognisable forms.

Tracking parameters on product and pricing pages

A Google Ads campaign, a LinkedIn link or an email add gclid, utm_source or other parameters. Google's own example is a URL followed by ?gclid=ABCD. Each combination is a distinct URL with identical content: a self-referencing canonical on the pricing page settles it without touching the campaigns.

Catalogue filters and sorting

An integrations page, a template library or a partner directory sorted by date or filtered by category produces variants. Google cites the sorting and filtering functions of a category page as a typical source of duplicates. Variants that add no different main content point to the default view as canonical.

Protocol and host variants

HTTP and HTTPS, with or without www. Google states it prefers HTTPS as canonical. A permanent redirect is the right tool here: these URLs have no reason to stay viewable.

Demo and staging environments

Google calls them “accidental variants”, such as a demo version left accessible to crawlers. It is common when a staging subdomain or a preview domain is not protected. It is also why Google recommends absolute URLs: a relative path on a crawlable environment points to the environment itself.

Cases outside the canonical's remit matter just as much. Two distinct pages in the same language targeting the same intent are not technical duplicates: that is SEO cannibalization, fixed by merging or editorial repositioning. And during a redesign, old URLs go through redirects, as covered in our SEO migration guide.

FR and EN versions: why each language keeps its own canonical

Because Google does not treat them as duplicates. Its documentation states that different language versions of a page are only considered duplicates when the main content is in the same language. A translated English page is therefore not a copy of the French one.

The instruction is explicit: when using hreflang, specify a canonical page in the same language. The classic mistake is pointing the canonical of /en/pricing to /tarifs, which amounts to asking Google not to show the English version. On this site, each article declares its own canonical and its counterpart in the other language, the mechanism detailed in our article on hreflang.

The errors that get a canonical tag ignored

An ignored canonical almost always signals a conflict between signals. Google documents several, each verifiable in the rendered code.

• Contradictory canonicals: one URL in the sitemap, another in the tag, a third in internal links. Google recommends linking internally to the canonical URL rather than a duplicate.
• A fragment as canonical: a URL with #section. Google generally does not support URL fragments.
• Noindex to arbitrate: Google advises against it within a single site, because it removes the page from Search entirely.
• Robots.txt to arbitrate: a blocked URL can stay indexed without its content, and Google can no longer read its tag.
• A canonical changed by JavaScript: Google recommends declaring it in the HTML source and making sure JavaScript does not change it.

When the gap between declared and selected canonicals affects whole templates, the cause is rarely a single page. A technical SEO project on indexing and URL signals fixes the template, and an SEO audit that prioritises bottlenecks tells you whether the issue deserves attention before the others.

About the author

Franck Gordi

Independent SEO and GEO consultant for B2B SaaS. He helps software companies become visible on Google and cited by AI engines.

LinkedIn →