Technical SEO · 17 Nov 2026 · 9 min read
Canonical tags explained: when and how to use them
A canonical tag tells search engines which address is the main version of a page. It is a hint, so it works best when the rest of your site agrees with it.
The short answer
A canonical tag is a link element in a page's head that names the preferred address for that content. Search engines use it to choose which of several similar URLs to index and show. Google treats it as a strong hint, not an order, so internal links, redirects and the sitemap should point to the same address.
What a canonical tag does
A canonical tag is a single line in the head of a page: a link element with rel set to canonical and an href holding the address you want treated as the main one. When several addresses show the same or very similar content, the tag says which one you would like indexed. People often call it rel canonical.
The need arises more often than most owners expect. One product can be reached by a category path, a search path and a tracking link. One blog post can load with and without a trailing slash. Each version is a separate URL to a crawler. Without guidance, a search engine has to guess which one to keep, and the guess may not be the one you print on business cards.
A canonical tag does three practical things. It gathers duplicate versions under one address so reports are easier to read. It reduces the chance that a messy variant, such as one carrying a campaign parameter, appears in results instead of the clean page. And it passes the signals from the variants towards the chosen page, so links to a variant are not wasted.
It does not hide a page, remove a page, or move visitors. Visitors still see whichever address they opened. If you need people and crawlers to land somewhere else, that is a redirect, covered below. For the wider picture of how pages get found and stored, see our guide on how search engines crawl, index and rank pages.
When a page needs a canonical tag
The honest answer is that every indexable page benefits from a self-referencing canonical tag, meaning the tag points at the page's own clean address. It costs nothing and protects against accidental variants. The cases below are where the tag does real work.
- Tracking and filter parameters. A link with utm parameters, a sort order or a session value loads the same page under a new address.
- Product variants. A shirt in four colours may have four addresses with near identical text. If the variants are not worth searching for separately, point them at the main product page.
- Printer friendly or AMP style copies, where a second version of the article exists on purpose.
- Syndicated content. If you publish an article on your site and a partner republishes it, the partner can point a canonical tag back to your original. That depends on their agreement; you cannot force it.
- Protocol and host variants, such as http and https, or with and without www. Redirects are the better fix here, with the canonical tag as a backup.
Pagination is a different case. Page two of a category is not a duplicate of page one, so pointing every page of a series at page one tells search engines to ignore the products listed on later pages. Give each page its own self-referencing tag unless you have a specific reason not to.
Two pages that compete for the same search are a related but separate problem. A canonical tag can merge them if they really are the same thing, but if they serve different needs, rewrite one of them. Our guide on keyword cannibalisation covers how to decide.
How to write a canonical tag correctly
Most canonical problems come from small slips, not from misunderstanding the idea. Work through these points when you set one up or review one.
- Put it in the head of the HTML, not the body. A tag placed in the body is ignored.
- Use one tag per page. Two tags that disagree, which happens when a theme and an SEO plugin both add one, leave the crawler to guess.
- Use the full address including https and the host name, not a relative path. Absolute addresses leave less room for error.
- Point to a page that returns a normal successful response. A canonical that leads to a 404, a redirect or a page marked noindex sends mixed messages.
- Keep the target in line with the content. Canonical tags suit pages that are the same or close to it. Pointing a detailed service page at the home page is a request Google is likely to ignore.
- Match the casing, trailing slash and parameter handling you use everywhere else on the site.
The tag can also be sent in an HTTP header, which is useful for files that have no HTML head, such as a PDF. Most small business sites will never need that, but it is worth knowing it exists if a developer mentions it.
If your site is built on a content management system, check whether it writes the tag for you before adding one by hand. Many do, and a hand written tag on top of a generated one is the usual source of duplicates.
Canonical tag, redirect or noindex: which one to use
These three tools solve different problems, and mixing them up causes the most damaging mistakes. The table sets out the difference.
| Tool | What it tells search engines | What visitors see | Use it when |
|---|---|---|---|
| Canonical tag | This page is a copy; prefer that address | The page they opened | Both versions must stay available to people |
| 301 redirect | This address has moved for good | The new address | The old address has no reason to exist |
| Noindex | Do not keep this page in search results | The page they opened | The page is useful to visitors but has no search value |
If an old address should simply stop existing, redirect it. Our explainer on 301 and 302 redirects shows which type to pick. If a page is a copy that still needs to be reachable, use a canonical tag. If a page should never appear in search at all, such as a thank you page, use noindex. Avoid combining a canonical tag with noindex on the same page, because the two instructions pull in opposite directions.
Why Google ignores your canonical tag
Google describes the canonical tag as a signal it weighs with others. It does not promise to follow it. When you see the message in Search Console that Google chose a different canonical from the one you declared, one of these causes is usually behind it.
The first is conflicting signals. Your tag says one address, but your internal links, your sitemap or your redirects point at another. Google tends to trust the address your whole site keeps pointing to. Fix the links and the sitemap, and the tag usually gets honoured. Our guide to internal linking is relevant here, but if you are short of time, start by making the sitemap list only the canonical addresses.
The second is content that is not really the same. If the two pages differ in substance, Google may decide they deserve separate treatment and keep both.
The third is a faulty target: a canonical that leads to an error page, a redirected address or a blocked page. If Google cannot fetch the target, the instruction cannot be used.
The fourth is a tag Google never saw. If a robots.txt rule blocks the duplicate, the crawler cannot read the tag on it. Blocking and canonicalising the same URL cancels the effect, which is why our guide to robots.txt warns against it. A tag added only by script after the page loads can also be missed or read late.
The fifth is simply that Google disagrees. If the chosen address is still a sensible one, you may not need to fight it. Check which address appears in results and whether it works for visitors before spending hours forcing the point.
How to check what Google has chosen
You can inspect a canonical in three places, and each answers a slightly different question.
View the page source in your browser and search for canonical. This shows what your site declares, including whether a plugin has added a second tag.
The URL Inspection tool in Google Search Console shows the user declared canonical beside the one Google selected. If they differ, you have found the page to investigate. Our beginner guide to Search Console explains how to open it. The page indexing report also lists pages excluded because Google picked another canonical, which gives you a list to work through.
A site crawl tool can list the canonical of every page at once and flag pages with missing, duplicate or conflicting tags. That is the quickest way to catch a template mistake that affects hundreds of pages.
When you review the results, look for patterns rather than single pages. A tag that points every page at the home page, or a tag that includes a staging address left over from the build, is a template fault that affects the whole site.
Mistakes that cause real damage
- Every page points to the home page. This usually comes from a default setting. It asks search engines to treat the entire site as one page.
- The staging address is canonical. After a site move, tags that still name the test site tell Google the real site is the copy.
- Canonical chains. Page A points to B, which points to C. Point everything at the final address.
- Canonical across languages. Translated pages are not duplicates of the original. Treat them with the hreflang approach instead.
- Different content, same tag. Pointing a long guide at a short summary of it can lead to the guide dropping out.
A weekly glance at the Search Console excluded pages list is enough for most small sites. If the numbers jump after a redesign or plugin update, check the tag first.
A simple routine for a small site
Start with the pages that matter: the home page, service pages and the top articles. Check each has one self-referencing canonical tag in the head. Then make sure the sitemap lists only those clean addresses. Next, search your own site for parameter versions and decide which ones need a tag pointing back to the clean page. Finally, check Search Console once a month for pages where Google chose a different canonical, and ask whether the difference is a problem or just a sensible choice.
If you would like a second pair of eyes on a site with many variants, such as a store with filters, our search engine optimisation service includes technical reviews of this kind. For a broader plan, read how to build an SEO strategy for a small business, and remember that the search intent behind a page, covered in our search intent guide, decides whether two pages are truly duplicates.
Related guides
- SEO services
- How search engines work crawling indexing ranking
- Seo strategy for a small business
- Search intent explained
- Robots.txt Explained: What to Allow and Block
- FAQ Sections for SEO: Real Questions and Markup
Questions and answers
What does a canonical tag do?
It names the preferred address for a page when several addresses show the same content. Search engines use it to decide which version to index and show. It does not move visitors or remove pages.
When do I need a canonical tag?
Use one on every indexable page as a self reference, and add pointing tags on duplicate or near duplicate versions such as parameter links, product variants and print copies. Pages with genuinely different content should not point at each other.
Why does Google ignore my canonical?
Google treats the tag as a hint. It may pick another address if your links, sitemap or redirects point elsewhere, if the target is broken, or if the pages are not really the same. Check URL Inspection in Search Console to see both choices.
Can I use a canonical tag instead of a 301 redirect?
Only if both addresses must stay open to visitors. If the old address has no reason to exist, a 301 redirect is the cleaner fix because it moves people and crawlers together.
Should every page have a self-referencing canonical tag?
It is good practice. A tag that points at the page's own clean address protects against accidental variants and is rarely harmful. Make sure the address is exactly the one you want indexed.
Tell us what you are building.
A short call, a written plan and a fixed quote. No obligation.