Most websites have duplicate pages without anyone deciding to create them. A product page reachable with three different URL parameters, the same blog post available with and without a trailing slash, a filtered version of a category page that Google finds on its own. None of this is a mistake exactly. It’s just how websites work. The canonical tag is the small piece of code that tells Google which one of those duplicates actually counts.
Quick summary, if you’re short on time:
- A canonical tag is a line of code that tells Google “this is the real version of this page, treat any similar URLs as copies of it.”
- Google has said directly that canonical tags are a strong hint, not a command. Google can and does ignore them if other signals disagree.
- The most common mistake isn’t missing canonicals. It’s canonical tags that point to the wrong page, a redirected page, or a page blocked from indexing.
- Search Console’s URL Inspection tool shows you both what you declared and what Google actually picked, and those two numbers disagreeing is a real warning sign.
- Every page should almost always have a canonical tag, even one pointing to itself. It’s cheap insurance against duplicate URLs you don’t even know exist yet.
Jump to a section:
- What a canonical tag actually is
- Why your site has duplicate URLs without you creating them
- A hint, not a command
- The mistakes that quietly cost rankings
- How to check your own site
- What this looks like in practice
- Using AI to sort out your own duplicate URLs
- Questions worth answering directly
What a canonical tag actually is
A canonical tag is a small piece of code in a page’s HTML that points to the URL Google should treat as the “master” version. It looks like this: a link element with rel="canonical" and an href pointing to the preferred URL. It sits in the page’s head section, invisible to a normal visitor.
The idea is simple. If three URLs on your site all show basically the same content, you pick one as the real one and tell Google, through the canonical tag, that the other two are just copies pointing back to it. Instead of splitting ranking signals across three weak versions of a page, they get combined onto the one you chose.
Why your site has duplicate URLs without you creating them
Nobody sits down and deliberately builds duplicate pages. They show up as side effects of how the web actually works. A tracking link with a UTM parameter on the end technically creates a new URL. A filter or sort option on a category page, like sorting products by price, often creates a new URL too. Some sites are reachable with and without “www,” or with and without a trailing slash at the end, and to a browser those look identical but to Google they’re technically different addresses.
Ecommerce sites get hit hardest, since a single product often has a URL for every category it’s filed under, plus separate URLs for every filter and sort combination a shopper might use, sometimes called faceted navigation. A service business isn’t immune either. A blog post syndicated to a partner site, a print-friendly version of a page, or a staging URL that accidentally got indexed can all create the same problem: multiple URLs, one real piece of content.
A hint, not a command
This is the part a lot of people get wrong. A canonical tag is not an instruction Google obeys automatically. Google’s own documentation is direct about this: canonical tags are treated as a strong signal, not a directive, and Google reserves the right to pick a different URL as canonical if other evidence points that way.
In practice, this usually only becomes a visible problem when your canonical tag disagrees with itself, like pointing to a page that redirects somewhere else, or a page that’s blocked from being indexed. When the signals conflict, Google has to guess, and its guess doesn’t always match what you intended.
This sits alongside two other technical checks worth doing together: whether Google’s mobile version of your page matches what you intended, and whether AI crawlers can even reach your site in the first place. Canonical tags, mobile rendering and crawler access are three separate signals, but they all decide the same thing: which version of your content Google actually trusts.
The mistakes that quietly cost rankings
The single biggest mistake isn’t forgetting canonical tags entirely, it’s canonical tags that point somewhere broken or contradictory. A canonical tag pointing to a URL that then redirects to a third URL creates a chain Google has to untangle, and it may just give up and pick something else. A canonical tag pointing to a page that’s also marked “noindex” sends two directly conflicting instructions on the same page, and Google has to choose which one to believe.
Missing self-referencing canonicals cause quieter damage. Even a page with no known duplicates benefits from pointing its canonical tag at its own URL, because it protects against duplicate versions you don’t know exist yet, like a new tracking parameter someone adds to a link next month. Paginated content is its own trap: each page in a series, page 2, page 3, and so on, should point its canonical tag at itself, not all funnel back to page 1, or Google may drop the deeper pages from the index entirely, along with whatever unique content they held.
Then there are the CMS-level accidents. A plugin or template bug that hardcodes the same canonical URL across every page on the site is more common than people expect, and it quietly tells Google that dozens or hundreds of different pages are all “copies” of one single page. This one is nasty precisely because nothing looks obviously wrong when you’re just browsing the site normally.
Two more worth knowing. A canonical chain, where page A points to page B, which itself points to page C, is a mistake on its own, even if every individual link looks fine. Point every canonical straight at the final URL, not at another canonical. And a canonical tag only works when the two pages are genuinely similar. Pointing it at a page with substantially different content is a signal Google is likely to just ignore, not a shortcut for merging two unrelated pages.
A short implementation checklist worth running through on any page: use the full, absolute URL, including https://, not a relative path. Use exactly one canonical tag per page, never two competing ones. Keep it in the page’s <head>, not the body, where crawlers won’t reliably read it. And don’t mix signals, a canonical tag pointing one way while an HTTP header or XML sitemap entry points another creates the exact conflict Google has to guess its way through.
How to check your own site
The fastest manual check: open any page, right-click, choose “View Page Source,” and search for rel="canonical". Confirm it points to the URL you’d actually want ranking, not a redirected or parameter-heavy version.
For a fuller picture, use the URL Inspection tool inside Google Search Console. It shows two separate fields worth comparing: “User-declared canonical,” which is whatever your page’s code says, and “Google-selected canonical,” which is what Google actually decided to use. When these two disagree, that’s Google directly telling you it didn’t trust or agree with your signal, and it’s worth investigating why.
To find issues across the whole site rather than one page at a time, open the Pages report in Search Console and look specifically for “Duplicate without user-selected canonical” and “Alternate page with proper canonical tag.” The first is Google telling you it found duplicates and picked a canonical itself, without you saying which one you wanted, worth reviewing every time it appears.
If you’d rather have this checked properly instead of digging through reports yourself, ask us for a canonical and duplicate URL audit as part of a free site review.
What this looks like in practice
Picture a Bengaluru interior design firm whose website lets visitors filter the portfolio page by room type: kitchens, bedrooms, living rooms. Each filter click updates the URL with a parameter, so the portfolio page effectively exists at a dozen different addresses, all showing overlapping content.
Without a canonical tag, Google might index several of these filtered URLs separately, splitting the ranking value of what should have been one strong portfolio page across many weak, near-identical ones. With a canonical tag on every filtered version pointing back to the plain, unfiltered portfolio URL, all of that value consolidates onto the one page that’s actually worth ranking.
Using AI to sort out your own duplicate URLs
Once you’ve listed out the URL variants a page actually has, working out which one should be canonical, and whether the ones you’ve picked are consistent with your redirects and indexing settings, is exactly the kind of structured check a prompt can help you think through.
“Act as a technical SEO helping me sort out canonical tags for a set of duplicate or near-duplicate URLs.
My URLs:
— List every variant of this page you know about: [paste the URLs, e.g. with/without www, with/without trailing slash, with tracking parameters, filtered versions]
— Which one do you currently treat as the “main” version, if any: [URL, or “not sure”]
— Do any of these redirect to another one already: [yes, list them / no / not sure]
— Are any of them marked noindex: [yes, which ones / no / not sure]
— Is this a single page, or a whole pattern, like every product or category page on the site: [single page / a repeating pattern]Task:
1. Tell me which URL should be canonical, and explain why in plain terms.
2. Flag any conflict in what I’ve described, like a canonical pointing to a redirected or noindexed page.
3. If this is a repeating pattern, describe the general rule I should apply across every instance, not just this one page.
4. Tell me what to check next in Search Console to confirm Google agrees with the canonical I’ve chosen.”
This is a reasoning aid, not a replacement for actually checking your site and Search Console. It can’t see your real URLs, redirects, or indexing settings, only what you describe. List every variant honestly, since a missed one is exactly the kind of gap that causes the mistakes covered above.
Questions worth answering directly
Does every page need a canonical tag?
Yes, as a general habit, even pages with no known duplicates. A self-referencing canonical costs nothing and protects against duplicate versions that might get created later without you noticing.
Can a canonical tag point to a page on a different domain?
Yes, this is called a cross-domain canonical, and it’s the standard way to handle syndicated content, telling Google the original source is the real version rather than the site republishing it.
What happens if my canonical tag points to a 404 or redirected page?
Google has to resolve the chain itself, and results are inconsistent. Always point canonical tags directly at the final, live URL, never at one that redirects or errors.
Is a canonical tag the same as a 301 redirect?
No. A 301 redirect actually sends visitors and bots to a different URL. A canonical tag lets both URLs stay live and reachable, it just tells Google which one to treat as the real one for ranking purposes.
Why would Google ignore my canonical tag?
Usually because other signals disagree with it, like internal links pointing to a different version, or the declared canonical page having weaker content than the one Google decided to use instead. Check the URL Inspection tool to see exactly what Google chose and why it might differ.
Should I ever avoid using a canonical tag?
Yes, when two pages aren’t actually similar. A canonical tag is meant to point at a near-duplicate, not to merge two genuinely different pages. If the content is substantially different, Google is likely to ignore the tag rather than follow it, so it isn’t a shortcut for combining unrelated pages.
This is the same underlying problem as keyword cannibalisation, multiple pages competing for the same ranking signal, and canonical tags are often the actual fix once you have diagnosed which pages are involved.
Where this actually pays off
Canonical tags are one of those unglamorous technical details that nobody notices when they’re working, and that quietly cap your rankings when they’re not. Getting them right doesn’t create new rankings out of nothing. It stops the rankings you’ve already earned from being split thin across duplicate URLs that were never supposed to compete with each other.
Not sure whether your site’s canonical tags actually say what you think they say? Share your site and we’ll check them against what Google is actually indexing, as part of every SEO engagement we run.