Canonical Tags and Duplicate Content: A Beginner Guide
beginner guideHey sister. A canonical tag tells Google which version of a page you prefer when several addresses show the same content. That is the cure for most duplicate content worries. This guide explains the tag, the myths around it, and a safe way to set it up.
- myth-free
- step-by-step fix
- Google’s own rules
// example data. Three addresses, one page, one preferred version.
What a Canonical Tag Does for You
A canonical tag names the preferred address when the same or very similar content lives at several addresses. Google treats it as a strong signal, not an order.
Say your sofa page opens at three different web addresses because of tracking codes and sort options. To you it is one page. To Google it is three addresses, and it has to guess which one to show. A canonical tag removes the guessing.
Google calls the process canonicalization: when several URLs serve the same or very similar content, you can indicate which one you prefer as the canonical URL. Its documentation lists redirects, the rel=”canonical” annotation and sitemap inclusion as ways to say so.
If you are brand new, read what is SEO and how Google search works first, because indexing is the stage where this matters.
The Duplicate Content Penalty Myth
Here is the myth that refuses to die. There is no “duplicate content penalty” for ordinary duplicates. Google’s starter guide says duplicate content isn’t a penalty, but it is inefficient and can confuse users. Calm down, put the kettle on, and fix it in an orderly way.
Where Duplicate Content Comes From
Most duplicates are accidents: tracking codes, sort filters, http and https versions, and printer-friendly pages. Rarely is it anyone copying anyone.
Common sources I see on small sites: product pages reachable under two categories, a homepage that opens with and without a slash, pages that open on both http and https, and tag or archive pages that repeat blog text. Each one gives Google extra addresses to crawl for no benefit.
Google also warns that if you do not specify a preference, it may choose a version it considers objectively the best version to show to users in Search. That might not be the one you wanted. So it pays to speak up.
?utm_source=newsletter
?sort=price-low
http:// and https://
one product, two paths
// example causes, from my own experience of small sites.
Three Ways to Name Your Preferred Page
Use a redirect when you are retiring a page, a canonical tag when both pages must stay, and the sitemap as a supporting hint. Google lists them from strongest to weakest.
Method 1: A Permanent Redirect
Google calls a permanent redirect a strong signal that the target should become canonical. Use it when you are getting rid of the duplicate. Visitors and Google both land on the right page. The details of redirect types are in 301 vs 302 redirects.
Method 2: The rel=canonical Annotation
Add a link element with rel=”canonical” to the head of the HTML page, or send an equivalent HTTP header, which works for files like PDFs. Google recommends choosing one of the two, not both. Use it when the duplicate must stay live, as with a sorted version of a product list.
Method 3: Sitemap Inclusion
Listing a URL in your sitemap is a weak signal, but it helps. Google still decides which pages are duplicates. Include only the canonical addresses. See robots.txt and sitemaps.
You can combine methods, which increases the chance that your preferred URL is chosen. Google says none of them are required, so think of them as strong suggestions.
How to Add a Canonical Tag Correctly
Put one canonical link in the head of each page, pointing to the full address of the preferred version. The page can point to itself too, which is common and harmless.
Google says to use absolute paths rather than relative ones with rel=”canonical”, because relative paths can cause problems, for example if a testing site gets crawled. So write https://www.example.com/sofas/corner-sofa/ in full.
On WordPress, SEO plugins add the tag automatically and let you override it per page. Check one page by viewing its source, or use the URL Inspection tool in Search Console to see which canonical Google chose. If you have not connected it, follow Google Search Console setup.
// source: Google Search Central. Bar lengths only show the order, not a measured score.
A Worked Example: A Sofa Shop in Manchester
A sofa shop in Manchester sells a corner sofa. The product opens at /sofas/corner-sofa/, but its newsletter link adds a tracking code, and the category page adds a sort option. Search Console shows several of these addresses.
The developer adds a canonical tag to each version pointing to the clean address with https and the full domain. The old http addresses get permanent redirects. The sitemap lists only the clean address. Internal links in menus and product grids use the clean address too, which Google says matters.
A week later the owner uses URL Inspection on a tracked address and checks which canonical Google chose. All data and names here are example data. If you want a professional to untangle this for you, my technical SEO work covers it, and my cheap SEO service for small business page shows the plans.
| task | method | status |
|---|---|---|
| tracked links | rel=canonical | done |
| http addresses | redirect | done |
| sitemap | clean URLs only | done |
| menu links | point to clean URL | done |
// example data, not a client result.
Canonical Mistakes and How to Fix Them
Most canonical problems come from mixed signals. Pick one preferred address and make every signal agree.
Google’s documentation lists mistakes worth avoiding. Here are the ones I meet most, with the fix.
- Using robots.txt to pick a canonical. Disallowed URLs may still be indexed. Fix: use a redirect or a canonical tag instead.
- Using noindex to pick a canonical. Noindex blocks the page entirely. Fix: keep the page indexable and point to the preferred one. See the noindex tag.
- Different canonicals from different methods. A tag says A, the sitemap says B. Fix: make them agree.
- Linking internally to duplicates. Fix: link to the canonical address everywhere. Read internal linking.
- HTTP leftovers. Google lists HTTP URLs in your sitemap or hreflang as a mistake. Fix: use https everywhere.
Another to watch: pages that compete for the same keyword are a different problem. For that, read keyword cannibalization.
Do
- Use absolute URLs in canonical tags
- Redirect pages you are retiring
- Link internally to the clean address
- Keep sitemap and tags in agreement
Never
- Block duplicates in robots.txt
- Use noindex to choose a canonical
- Canonicalize every page to the homepage
- Fear a made-up duplicate penalty
How to Check Which Version Google Picked
Use the URL Inspection tool in Search Console. It shows the canonical you declared and the one Google selected.
Paste a duplicate address into the inspection bar and read the indexing section. If your declared canonical and Google’s selected canonical differ, look for conflicting signals: a redirect pointing one way, a sitemap listing another, internal links to a third. Fix the odd one out and wait. Google’s starter guide says changes can take anywhere from hours to several months, so recheck in a few weeks.
For a whole-site review, run a basic SEO audit and sort the issues you find by how many pages they touch.
A Simple Routine to Keep Duplicates Under Control
Check three things every few months: your sitemap, your internal links and your Page indexing report. Small, regular checks beat one giant cleanup.
Duplicates creep back in. A new plugin adds archive pages, a marketing tool adds tracking codes, a redesign changes the address format. None of it is dramatic, but it adds up.
Every Few Months
Open your sitemap and scan it. Does it list only the clean addresses you want found? Click a few menu and product links and look at the address bar. Do they point to the clean version? Then open the Page indexing report in Search Console and look for pages Google chose to treat as alternates. Many of those are expected and harmless. The ones to investigate are important pages that Google skipped in favor of some other address.
When You Launch Anything New
Before a new section goes live, ask one question: could this content be reached at more than one address? If yes, decide the preferred one now, add the tag, and redirect the rest. Doing it at launch takes minutes. Doing it after a year of links and bookmarks takes an afternoon and a headache.
My honest opinion: most small sites need a handful of canonical fixes, not a campaign. Fix the obvious ones, keep your signals consistent, and spend the saved time writing something worth reading.
Canonical Tags and Duplicate Content: Questions Answered
Does duplicate content get a penalty?
Not for ordinary duplicates. Google’s starter guide says duplicate content isn’t a penalty, but it is inefficient and can confuse users. Copying other people’s content is a different matter, and the guide says not to do it.
Is the canonical tag a command?
No. Google describes redirects, canonical annotations and sitemaps as signals and preferences, and says none of them are required. Google may choose a different version it considers best for users.
Should a page canonicalize to itself?
Yes, that is fine and common. A self-referencing canonical tells Google that the clean address is the one you prefer, even when tracking or parameter versions appear.
Redirect or canonical, which is better?
Use a permanent redirect when you are retiring the duplicate, since Google calls it a strong signal. Use a canonical tag when both addresses must stay reachable, for example sorted product lists.
Can I use relative URLs in canonical tags?
It is better not to. Google recommends absolute paths with rel canonical, because relative paths can cause problems, for example if a testing site gets crawled. Write the full https address.
Send Me Your URL. I’ll Tell You What I’d Fix.
One WhatsApp message with your website address is enough for a first honest look.
Sources and Last Check
- Google Search Central: Consolidate duplicate URLs (methods, absolute paths, mistakes)
- Google Search Central: SEO Starter Guide (duplicate content is not a penalty)
- Google Search Central: Introduction to robots.txt (blocked URLs can still be indexed)
// last checked 11 Nov 2026. The Manchester sofa shop is example data. Opinions are marked as opinions.