What Is a Canonical Tag? Your Duplicate Content Referee

Jul 23, 2026 · SEO Glossary

Run an online store for five minutes and you'll discover an awkward truth: your site is generating duplicate pages without your permission. The same blue kettle lives at /kitchen/kettles/blue-kettle/, at /sale/blue-kettle/, at /blue-kettle/?color=blue&utm_source=newsletter, and maybe at the same URL with a session ID stapled on. Four addresses, one kettle. To a human, obviously the same page. To a search engine, four separate URLs competing with each other.

The canonical tag is how you referee that competition. What is a canonical tag? It's a snippet in a page's <head> — <link rel="canonical" href="https://example.com/kitchen/kettles/blue-kettle/"> — that declares which URL is the authoritative, "real" version of this content. It tells search engines: index and rank that one; treat the rest as copies and consolidate their signals onto it.

Why duplicates happen to good sites

Nobody plans duplicate content. It emerges from the mechanics of how sites work:

URL parameters: tracking tags (?utm_campaign=spring), sort orders (?sort=price), filters, session IDs — each combination is technically a distinct URL serving near-identical content.
Protocol and host variants: http:// vs https://, www vs bare domain, trailing slash vs none. Sloppy server config can leave several combinations all returning 200.
Faceted navigation: e-commerce filters can mint thousands of URL combinations from one category.
Printer/AMP/mobile variants, paginated comment pages, and staging copies that got indexed.
Syndication: your article republished on Medium or a partner site — same words, different domain.

The cost of leaving this unmanaged: ranking signals split across variants (three backlinks to one URL, five to another, none reaching full strength), Google picking its own favorite version — possibly the ugly parameterized one — and crawlers burning crawl budget on the same kettle forty times.

What is a canonical tag telling Google — and does Google obey?

Here's the nuance that trips up even experienced people: the canonical tag is a hint, not a command. Google weighs your declared canonical alongside other signals — internal links, sitemap inclusion, redirects, which version has backlinks, content similarity — and usually honors it. But when your hint contradicts the other evidence, Google overrules you and picks its own canonical. Search Console tells you when this happens: the URL Inspection tool shows both "user-declared canonical" and "Google-selected canonical," and the page indexing report has a whole category for "Duplicate, Google chose different canonical than user."

This design makes sense once you see canonicals as one witness among several. If your tag says /blue-kettle/ is canonical but your entire menu links to /kitchen/kettles/blue-kettle/ and your XML sitemap lists a third variant, you've given contradictory testimony. Consistency is what makes your canonical stick: tag, internal links, and sitemap all pointing at the same URL.

Contrast this with a 301 redirect, which is a command — the old URL stops being reachable at all. The decision rule: if users never need the duplicate URL, redirect it. Canonicals are for when the variants must stay accessible (filtered views, tracking URLs, print versions) but shouldn't be indexed separately.

Placing canonicals correctly

Use absolute URLs. href="https://example.com/page/", never href="/page/" — relative canonicals invite protocol and host ambiguity.
Self-canonicalize by default. Every indexable page should carry a canonical pointing at itself. It costs nothing and preempts parameter-based duplicates automatically: /page/?utm_source=x inherits the tag pointing at clean /page/.
One canonical per page. Two canonical tags with different targets means Google ignores both. This happens more than you'd think — usually an SEO plugin and a theme both injecting one.
Point at live, indexable pages. A canonical to a 404, a redirected URL, or a noindexed page is contradictory testimony again.
Cross-domain works. Syndication partners can canonical their copy back to your original — the polite arrangement worth requesting whenever you republish.
For non-HTML files (PDFs, say), send it as an HTTP header: Link: <https://example.com/doc/>; rel="canonical".

The mistakes that make canonicals backfire

A few patterns I've pulled out of real audits, each one quietly sabotaging a site:

Canonicalizing everything to the homepage. A misconfigured template stamps href="https://example.com/" on every page — effectively asking Google to deindex the entire site. (Google usually recognizes the error and ignores it, but you're gambling.)
Canonicalizing paginated pages to page one. Page 2 of a category isn't a duplicate of page 1 — it has different products. Each paginated page should self-canonicalize; pointing them all at page 1 can orphan everything deeper.
Pointing at the redirected variant. Site moved to HTTPS, but canonicals still say http:// — every page now hints at a URL that 301s away. Chain confusion.
Canonical vs. noindex contradiction. A page that's canonicalized to another URL and marked noindex sends mixed signals about whether its equity should consolidate or vanish.
Environment leakage. Production pages canonicalizing to staging.example.com after a careless deploy. Genuinely deindexes things.

Auditing your canonicals

Spot-check important pages by viewing source and searching for rel="canonical", and use Search Console's URL Inspection to compare declared vs. Google-selected canonicals on your money pages. For full coverage, crawl the site: run an automated site audit and review every page's canonical target — flagging missing tags, canonicals pointing at redirects or 404s, and clusters of pages canonicalizing somewhere unexpected. The homepage-canonical template bug alone justifies the check; it's the kind of error that's invisible in the browser and catastrophic in the index.

Canonicals are unglamorous plumbing, and that's the compliment. Set them consistently, keep them agreeing with your links and sitemap, and duplicate content stops being a thing you think about.

Frequently Asked Questions

Is a canonical tag the same as a 301 redirect?

No. A 301 physically sends users and crawlers to the new URL — the old one stops being viewable. A canonical tag leaves both URLs accessible to users but tells search engines which one to index. Use a redirect when the duplicate URL has no reason to exist; use a canonical when users still need the variant, like filtered or tracking-parameter URLs.

Why did Google choose a different canonical than I declared?

Because canonical tags are hints, and your other signals disagreed. Google weighs internal links, sitemap entries, backlinks, and redirects alongside your tag — if your menu links and sitemap point at a different variant than the tag names, Google may trust them over you. Align tag, internal links, and sitemap on the same URL and your declaration will usually stick.

Should every page have a canonical tag?

Every indexable page benefits from a self-referencing canonical — it costs nothing and automatically neutralizes parameter-based duplicates like UTM-tagged URLs. Pages that are variants of another page should point their canonical at the primary version instead.

Do canonical tags pass link equity?

Yes — when Google honors the canonical, ranking signals from the duplicate URLs, including their backlinks, are consolidated onto the canonical version. That consolidation is the main SEO point of the tag: instead of three weak variants splitting the votes, one strong URL collects them.

Try WebsiteChecker.Tech Free

Run a free technical SEO audit on any website. Get a client-ready report in minutes.

Start Free Scan