One page can be reachable at several web addresses without anyone planning it. This guide explains what that does in search, how a canonical tag helps and how to find and fix duplicates.
Duplicate content is the same or very similar content available at more than one web address. Google says some duplication is normal and is not a spam violation, but it still has to choose one address to show. A canonical tag names your preferred address, and a redirect is stronger when the old address should stop being used.
What duplicate content is
Duplicate content is the same, or very nearly the same, content available at more than one web address. It can sit on one site or be spread across several. Most of it is created by accident, by the way a site is built, not by anyone copying anything.
Google's documentation on canonicalisation names the usual sources: regional versions of a page in the same language, separate mobile and desktop versions, the http and https versions of a site, sorting and filtering on listing pages, and a test copy of the site left open to crawlers. Small business sites meet a few more.
| Cause | Example | Usual fix |
|---|---|---|
| Protocol and host variants | http and https, or with and without www, all load the same page | Redirect every version to one |
| Trailing slash and letter case | /services and /services/ both load | Choose one form and redirect the other |
| Tracking or sort parameters | The same page with a campaign tag or a sort order added to the address | Canonical tag pointing to the clean address |
| Print or separate mobile versions | A second address built only for printing | Canonical tag pointing to the main page |
| A test copy left open | A staging site that anyone can reach | Put it behind a password |
| Archive pages in a CMS | One post listed under category, tag, author and date addresses | Decide which archives deserve to be indexed |
| Near-identical pages | Service pages that differ only by a place name | Merge them, or write each one properly |
| Copied text | Supplier descriptions used on many stores | Rewrite the pages that matter most |
Does Google penalise duplicate content?
Not in the way most people fear. Google's documentation says that some duplicate content on a site is normal and does not break its spam policies. Google has also said that duplicate pages are not grounds for action unless the duplication looks designed to deceive people or manipulate results.
What happens instead is quieter. Google groups pages whose main content is very similar and picks the one it judges most complete and useful as the canonical version. That page is the one shown in results. Signals such as links to the other versions are consolidated onto it, and the duplicates are crawled less often.
That still causes three practical problems for a site owner. The address Google picks may not be the one you wanted visitors to land on. Links to your content can be split between versions when Google does not recognise them as the same page. And crawling time is spent on copies instead of on new pages. None of these is a penalty, and all are worth fixing.
Copying other sites' content in bulk to manipulate results is a different matter. That falls under Google's spam policies and is not what this guide is about.
What a canonical tag is
A canonical tag is a line in the head section of a page that names the preferred address for that content. Technically it is a link element with the relation "canonical" and the full address of the preferred page. When several addresses show the same content, the tag tells search engines which one you want treated as the main version.
It is a signal, not an order. Google's documentation calls a canonical tag a strong signal that the named address should be canonical, and a redirect is also a strong signal. Listing an address in your sitemap is a weak one. The signals add up, so use more than one and make sure they agree. Google still compares the pages itself and may choose a different address.
Google's guidance on writing the tag comes down to a short list.
- Use the full address, including https and the domain, not a relative path.
- Put a tag on the preferred page too, pointing to its own address. This is called a self-referencing canonical.
- Use one method for a page: the link element in the HTML or an HTTP header. Google recommends choosing one, because using both is more likely to go wrong. The header version also works for files that are not web pages, such as PDFs.
- The link element is read only inside the head of the page.
- Link to the canonical address from other pages of your site, and list canonical addresses in the sitemap.
- Do not send conflicting signals, such as one address in the sitemap and a different one in the tag.
None of this is mandatory. Without a tag, Google chooses for itself, and on a small, tidy site that is often fine. The tag earns its place when you have a clear preference and the pages are giving mixed signals.
Canonical tag or 301 redirect?
Use a redirect when the old address should stop being used. Use a canonical tag when both addresses must stay reachable. A visitor who follows a redirect ends up on the new address. A visitor on a page with a canonical tag stays where they are, and only the search engine is told which address you prefer.
| Situation | Use | Why |
|---|---|---|
| Site moved from http to https | 301 redirect | Nobody should land on the old version |
| Both www and non-www load | 301 redirect | One host name should serve the site |
| An old page replaced by a new one | 301 redirect to the closest match | Visitors and links follow the move |
| The same page reached with tracking tags | Canonical tag to the clean address | The tagged links must keep working |
| Sort or view options visitors use | Canonical tag to the main listing | The option stays usable |
| A print version | Canonical tag to the main page | The print address stays reachable |
Send each old address to its closest equivalent. Redirecting everything to the home page, or to a page on another subject, helps nobody and can leave visitors as lost as an error page would. The guide to 404 errors and broken links covers redirect rules and what a useful error page looks like.
How to find duplicate pages
A small site can be checked in an afternoon with free tools. Work through these steps in order.
- Check address variants by hand. Type the http, https, www and non-www forms of your home page and of one inner page, with and without a trailing slash. Each should end up at a single address.
- Open the Page indexing report in Search Console. Statuses such as "Duplicate without user-selected canonical" and "Duplicate, Google chose different canonical than user" list pages that Google treats as copies of another page.
- Inspect a key page with the URL Inspection tool. It shows the canonical you declared and the canonical Google selected. When they differ, Google disagrees with your preference, or does not see the two pages as similar enough to be duplicates.
- Crawl the site with a crawler and sort the results by page title and main heading. Identical titles and headings are the quickest sign of pages that repeat each other.
- Search Google for a distinctive sentence from a page, in quotation marks. If other addresses return the same text, note them. This is a spot check, not a full audit.
- Compare the addresses in your sitemap with the ones a crawl finds. Addresses that appear in one list but not the other often point to variants.
If many pages are affected, or you cannot see why, the signs listed in the guide to when a site needs a technical SEO audit help you decide whether to bring in help.
How to fix duplicate content
Fix the cause in the site's set-up first, and the individual pages second.
- Decide the preferred form of every address: one protocol, one host name and one rule for trailing slashes.
- Redirect every other form to it with a permanent redirect, set once for the whole site.
- Put a self-referencing canonical tag on every page that should be indexed.
- For pages that must stay reachable under extra addresses, such as links carrying campaign tags, point the canonical tag to the clean address.
- Merge near-duplicates. Two pages that answer the same question should become one, with the old address redirected. A website content audit gives you a page-by-page way to decide what to keep, merge or remove.
- Give pages that differ only by a place name or product name enough content of their own to justify existing, or fold them together.
- Rewrite copied supplier or manufacturer text, starting with the pages that matter most to the business.
- Tidy internal links and the sitemap so that they use only the preferred addresses.
- Check Search Console again after Google has had time to recrawl. Statuses change only when Google revisits the pages.
Keep a note of the rules you chose. The next developer, plugin or redesign will otherwise undo them without knowing.
Online stores and CMS sites
Content management systems and online stores create duplicates faster than hand-built sites. A blog may list each post under category, tag, author and date archives. A store adds filters, sort orders and product variants, and one product can end up with a dozen addresses.
The details depend on the platform, and many themes and plugins set canonical tags automatically, sometimes wrongly. Check what yours outputs on a sample of pages instead of assuming. The guide to ecommerce SEO audit problems deals with filter addresses, variants and supplier text in detail, so this guide leaves those cases there.
Canonical tag mistakes to avoid
Most canonical problems come from templates, plugins or well-meant shortcuts. These are the ones Google's own guidance warns about, or that follow from it.
- Using robots.txt to deal with duplicates. A blocked page cannot be read, so Google cannot see its canonical tag. Google advises against robots.txt for canonicalisation.
- Using a noindex tag to choose between versions. It removes the page from search instead of consolidating it. A canonical tag is the preferred tool.
- Putting more than one canonical tag on a page. Google has said that when there are several, they are all ignored.
- Pointing to an address that does not exist, redirects elsewhere or carries a noindex tag.
- Pointing page two of a series back to page one. They are different pages, not duplicates.
- Pointing every page to the home page, which is often a template setting nobody noticed.
- Using a relative path, or placing the tag outside the head.
- Letting a script change the canonical after the page loads. Google advises setting it in the HTML source.
- Giving different answers in the tag, the sitemap and the internal links.
A canonical tag also does not repair thin or weak pages. If a page has little to say, the answer is better content or a merge, not a tag. Sites that want help with this kind of clean-up can read about Liftup's SEO service.
Questions and answers
What is duplicate content?
Duplicate content is the same or very similar content available at more than one web address, on one site or across several. Most of it is accidental, caused by address variants, parameters, archives or copied supplier text. Google chooses one version to show and treats the others as copies.
Does Google penalise duplicate content?
Google's documentation says some duplication on a site is normal and not a breach of its spam policies. The real costs are quieter: the wrong address may be shown, links can be split between versions, and crawling time is spent on copies. Deliberate copying to manipulate results is a different case.
What is a canonical tag?
A canonical tag is a line in the head of a page that names the preferred address for its content. It is a strong signal to Google, not a command, so Google may still choose another address. Use the full address and keep the sitemap and internal links consistent with it.
Do I need a canonical tag on every page?
None is required, and Google will pick a version itself if you add nothing. Google does advise placing a self-referencing canonical on the preferred page. Doing so on every indexable page is a simple habit that removes doubt about which address you mean.
How do I find duplicate pages on my site?
Check the http, https, www and non-www forms of your pages, then read the Page indexing report and the URL Inspection tool in Search Console. A crawl sorted by page title also shows repeated pages quickly. Compare your sitemap with what a crawler finds to spot stray variants.
How do I fix duplicate content?
Redirect old or variant addresses to one preferred address, add self-referencing canonical tags, and point extra reachable addresses to the clean one. Merge pages that answer the same question and rewrite copied text. Then keep internal links and the sitemap consistent with your choices.
Should I use a canonical tag or a redirect?
Use a redirect when the old address should stop being used, such as after a move to https. Use a canonical tag when both addresses must stay reachable, such as links with tracking tags. Both are strong signals, and they should never contradict each other.