Should you really place a canonical tag on every page, even without duplication?

Why does Google emphasize self-referential canonicals?

The recommendation from John Mueller may seem counterintuitive. If a page has no duplicate, why declare a canonical pointing to itself? The answer lies in how Googlebot discovers and indexes URLs.

A single page can be accessible via multiple URL variants: with or without a trailing slash, with tracking parameters (utm_source, etc.), in HTTP vs HTTPS, or with session parameters. Without an explicit directive, Google must guess which version to index. The self-referential canonical tag eliminates this ambiguity by stating: "here is the official version."

What is the difference with no canonical tag?

Without a canonical, Google applies its own algorithms to choose the canonical URL. This selection can differ from your preferences, especially if your site generates unintended URL variants.

For example, a filtering system may create /products?sort=price and /products?sort=name. Without a canonical, Google might index a filtered version instead of your main page. The self-referential canonical on /products forces Google to consider that version as the primary reference, even if variants exist.

Does this approach cover all scenarios?

No. The self-referential canonical works for unintentional or technical duplicates, not for intentionally duplicated content like identical product listings across multiple categories.

In these cases, you need to designate a master page and point the duplicates to it. Mueller's recommendation concerns unique pages that might be discovered under multiple variant URLs, not strategies for consolidating duplicate content.

• Each unique page should have a canonical tag pointing to its own clean URL
• This practice clarifies the preferred URL even without apparent duplication
• URL variants (parameters, trailing slash) are managed by the canonical
• True content duplicates require a strategy for canonical pointing to the master page
• Google uses this directive as a strong signal, not as an absolute directive

Is this recommendation really a priority?

Let's be honest: the self-referential canonical is a good defensive practice, not a direct ranking factor. On a well-configured site without wild URL parameters, the impact is marginal. Google usually identifies the correct version.

Where it becomes critical is on e-commerce sites, platforms with filters, or any system generating dynamic URL variants. I have seen cases where Google indexed hundreds of filtered pages because no canonical indicated the main version. In these contexts, Mueller's recommendation is not advice; it is a necessity.

What gray areas should you watch out for?

Mueller talks about "clarifying the preferred URL," but Google never guarantees to respect the canonical 100%. It's a strong signal, not a directive. If Google detects that your canonical points to a page significantly different from the crawled content, it may ignore it.

Another point is cross-domain canonicals. If you duplicate content across multiple sites (syndication, white-label), the canonical should point to the original source. However, Google may decide that another version is more relevant for certain queries. [To verify] in your crawl logs to see which URL Google actually indexes.

When does this rule become counterproductive?

Watch out for poorly calibrated automatic implementations. I have seen CMSs add self-referential canonicals with hard-coded absolute URLs, creating issues in staging or after HTTPS migration. If your canonical points to http:// while the site is on https://, you create a contradiction.

Another trap: paginated pages. Placing a self-referential canonical on page=2, page=3, etc. is correct only if each pagination page has unique and indexable content. Otherwise, all paginated pages should canonically point to page=1 or use rel=prev/next (although Google has deprecated this signal).

How to correctly implement this recommendation?

First step: audit your current URLs. Use Screaming Frog or a crawler to identify pages without a canonical tag. Also, check those that have a canonical pointing to a different URL without valid reason.

The technical implementation depends on your stack. On WordPress, Yoast and RankMath automatically add self-referential canonicals. On custom CMS or static sites, you need to manually implement it in the <head>: <link rel="canonical" href="ABSOLUTE_CLEAN_URL" />. Always in absolute, never in relative.

What critical errors should you avoid?

Do not let your canonicals point to URLs with tracking parameters. If your page is /product-x, the canonical should point to /product-x, not to /product-x?utm_source=email. Google may ignore the canonical if it contains irrelevant parameters.

Another common mistake: chained canonicals. Page A canonically points to B, which canonically points to C. Google generally follows only one hop, not more. Ensure that each page canonically points directly to the final version, without intermediaries.

How to check that everything works after deployment?

Use Google Search Console: section "Coverage" then "Excluded". If you see pages marked "Another page with appropriate canonical tag", it's a good sign — Google respects your directives. If strategic pages appear here, check their canonical.

Complete with the URL inspection tool. Paste a URL, look at "User-defined canonical URL" vs "Canonical URL selected by Google". If they differ, investigate why. Server logs and crawl data will tell you which version Google actually indexes.

• Add <link rel="canonical" href="ABSOLUTE_URL" /> in the <head> of each unique page
• Use clean absolute URLs (https, without unnecessary parameters, the preferred version of the trailing slash)
• Check that paginated or filtered pages correctly canonicalize (to main page or to themselves according to strategy)
• Audit Search Console to detect discrepancies between declared canonical and canonical retained by Google
• Test in staging before global deployment, especially after migrations or architecture changes
• Document your canonical strategy to avoid regressions during CMS updates