Canonical Tags: Telling Google Which Page Wins

The same page on your site probably lives at more than one address. The www version and the bare-domain version. The HTTP and the HTTPS. The clean URL and the one with a ?ref=newsletter tag stuck on the end. To you they are one page. To Google they can look like several near-identical copies, and that forces a question: which one should rank?

You can answer that question yourself with a canonical tag, or you can let Google guess. Most of the time you want to answer it.

What Counts as Duplicate Content
The Canonical Tag Is a Hint, Not a Command
How to Tell Google Which Page Wins
Does Duplicate Content Hurt Rankings?
Common Canonical Mistakes
Frequently Asked Questions

What Counts as Duplicate Content

Brass-SEO uses a simple definition that matches Google's: duplicate content is the same or very similar content reachable at more than one URL. It is far more common than most owners realize, and almost never on purpose.

The usual sources are mechanical. A store generates a new URL for every filter and sort option. A blog serves the same post under several tag and category paths. A site answers on both http:// and https://, and on both the www and non-www host. Tracking parameters from email and ads append strings that create a fresh URL for the same page. Each variation is a separate address pointing at content Google has already seen elsewhere.

None of this is sneaky, and Google knows it. The full set of specifications behind canonical handling lives in the canonicalization and duplicate content research topic, and the mechanics of how crawlers find all those URLs are covered in the crawling and indexing research topic. The job is not to hide the duplicates. It is to tell Google which version is the real one.

The Canonical Tag Is a Hint, Not a Command

Here is the detail that trips up even experienced site owners. Brass-SEO stresses it because it changes how you work: Google states that "indicating a canonical preference is a hint, not a rule," and that Google "may choose a different page as canonical than you do."

A canonical tag is a <link rel="canonical" href="..."> element that names the preferred URL for a piece of content. You are telling Google, "of all the addresses where this content appears, treat this one as the original." Google takes that as one signal among several. It also weighs your redirects, your internal links, which URL sits in your sitemap, and its own preference for HTTPS. If those signals point one way and your canonical tag points another, Google can side with the signals.

So the canonical tag works best when the rest of your site agrees with it. Point your internal links at the canonical URL. List the canonical URL in your sitemap. Make the tag the loudest voice in a chorus, not a lone vote.

How to Tell Google Which Page Wins

Brass-SEO recommends matching the method to the situation, because Google supports several and treats them with different weight. Google lists redirects, the rel="canonical" link element, a rel="canonical" HTTP header, and sitemap inclusion, and it notes the methods "can stack and thus become more effective when combined."

Use a 301 redirect when a duplicate should not exist on its own, like forcing http:// to https:// or www to non-www. A redirect is the strongest signal because it removes the duplicate entirely. Use the rel="canonical" link element when both URLs need to stay reachable, such as a product page that must remain live under several filter paths. Google accepts that element only when it appears in the page <head>, and it asks for absolute URLs rather than relative ones. The HTTP header version does the same job for non-HTML files like PDFs. Sitemap inclusion is the weakest of the signals, useful as reinforcement rather than the sole instruction.

For most small sites, two moves cover the ground. Redirect the protocol and host variants once at the server, and set a self-referencing canonical on each real page so every URL declares itself.

Does Duplicate Content Hurt Rankings?

This is where Brass-SEO calms a common fear. Google is explicit that "some duplicate content on a site is normal and it's not a violation of Google's spam policies." There is no duplicate-content penalty for the ordinary cases described above.

The real cost is quieter than a penalty. When several URLs hold the same content, Google picks one to treat as canonical and concentrates its evaluation there. Links and relevance signals that should reinforce a single strong page get spread across the copies instead. Pick the wrong canonical, or let Google pick for you, and the version that ranks may not be the one you would have chosen. When Google reassigns which URL is canonical, the page that shows in results can shift with it, one of the causes explored in the post on why rankings bounce around. Consolidating the duplicates pools those signals back onto one page, which is the version most likely to rank well. The broader picture of how Google weighs these signals sits in the ranking signals research topic.

Deliberate, large-scale copying is a different matter and can fall under spam policies. Ordinary technical duplication is a housekeeping task, not a violation.

Common Canonical Mistakes

Brass-SEO sees the same handful of errors again and again, and each one quietly sends the wrong page to the front.

Canonical pointing at the wrong page. A template bug that makes every page canonicalize to the homepage tells Google to drop all your other pages. Check that each page points to itself or its true original.
The tag placed outside the head. Google accepts the rel="canonical" link element only inside the <head>. A canonical injected into the page body by a script is ignored.
Relative URLs. A relative path can resolve to the wrong address. Google asks for absolute URLs, with the full https:// and domain.
Canonical fighting noindex. Telling Google a page is canonical while also telling it not to index that page sends contradictory instructions. Pick one.
Multiple canonical tags on one page. RFC 6596, the open standard that defines the tag, says to specify only one canonical relation per resource. Two tags let Google ignore both.

Frequently Asked Questions

Does Google penalize duplicate content?

Generally no. Google states that "some duplicate content on a site is normal and it's not a violation of Google's spam policies." The issue with duplicates is consolidation, not punishment. Google picks one URL as canonical and evaluates it, so without guidance it may rank a version you did not intend. Large-scale deliberate copying is a separate matter that can fall under spam policies.

Will Google always follow my canonical tag?

No. Google is clear that a canonical declaration is "a hint, not a rule" and that it "may choose a different page as canonical than you do." It weighs your tag against redirects, internal links, sitemap inclusion, and its HTTPS preference. To make your choice stick, align those other signals with the canonical URL rather than relying on the tag alone.

What is the difference between a canonical tag and a 301 redirect?

A 301 redirect sends every visitor and crawler from one URL to another, so the duplicate stops existing as a separate page. A canonical tag keeps both URLs reachable but tells Google which one to treat as the original. Use a redirect when a duplicate should not stand on its own, like protocol or host variants. Use a canonical when both URLs need to stay live, like filtered product pages.

How do I fix www and non-www showing as duplicates?

Pick one as your canonical host and 301-redirect the other to it at the server level, then do the same for HTTP to HTTPS. That single set of redirects resolves the most common duplication for a small site. Add a self-referencing canonical tag on each page so every URL also declares its preferred address.

Duplicate URLs are normal. Leaving the choice of which one ranks up to Google is the avoidable part. Set your canonicals, align your signals, and run a free first analysis to see which of your pages Google is actually indexing.