What Are Canonical Tags and How Do They Work?

A canonical tag tells search engines which URL you consider to be the main version of a page when the same, or very similar, content can be accessed through more than one URL.

The tag is placed in the <head> section of a webpage and normally looks like this:

<link rel="canonical" href="https://www.example.co.uk/preferred-page/" />

Its purpose is to help search engines consolidate duplicate URLs, understand which version should normally appear in search results and avoid splitting signals between several versions of essentially the same page.

Canonical tags are particularly useful on ecommerce websites, large content websites, sites that use URL parameters and websites where the same page can be reached through several different URL formats.

Importantly, a canonical tag is a strong signal rather than an absolute instruction. A search engine can select a different canonical URL if other signals suggest that another page is a more appropriate representative version.

Pro Tips – Get Your Canonical Tags Right

• Use Self-Referencing Canonicals – Important indexable pages should normally point to their own preferred URL.
• Choose One Preferred URL – Make it clear which version of duplicate or very similar content should be treated as the main page.
• Use Absolute URLs – Include the full HTTPS URL in your canonical tag rather than relying on a relative path.
• Keep Signals Consistent – Internal links, XML sitemaps, redirects and canonical tags should all support the same preferred URL.
• Check For Duplicate Content – Identify pages and URL variations that may compete or send conflicting signals to search engines.
• Avoid Broken Canonicals – Never point a canonical tag towards a 404 page, inaccessible URL or removed content.
• Do Not Canonicalise Unrelated Pages – Canonical tags should be used for duplicate or substantially similar content, not simply pages covering related topics.
• Use HTTPS Canonicals – If HTTPS is your preferred version, make sure canonical tags do not accidentally point back to HTTP URLs.
• Check Ecommerce Filters Carefully – Some filtered URLs may deserve their own indexable page rather than being canonicalised to the main category.
• Audit Canonicals Regularly – Check for missing tags, multiple canonicals, redirect chains and conflicting canonical destinations after website changes.

What Is a Canonical Tag?

A canonical tag, sometimes called a canonical link element or rel="canonical", is a piece of HTML that identifies the preferred URL for a page.

For example, imagine the same product can be reached through these URLs:

https://www.example.co.uk/product/blue-chair/

https://www.example.co.uk/product/blue-chair/?colour=blue

https://www.example.co.uk/product/blue-chair/?utm_source=email

If the first URL is the version you want treated as the main page, the other versions could contain:

<link rel="canonical" href="https://www.example.co.uk/product/blue-chair/" />

This tells search engines that although the other URLs are accessible, the clean product URL is the preferred version.

Some duplicate URLs are perfectly normal. Filtering systems, tracking parameters, print versions, product options and technical website configurations can all produce multiple URLs without anybody deliberately creating duplicate pages.

The objective of canonicalisation is not simply to remove duplicates. It is to make the relationship between those URLs clear.

If duplicate pages are becoming a wider problem on your website, my guide to duplicate content explains why multiple versions can appear and what to do about them.

How Do Canonical Tags Work?

When a search engine discovers several URLs containing identical or substantially similar content, it can group those URLs together and select one as the representative page.

The canonical tag is one of the signals it can use when making that decision.

For example:

URL A

https://www.example.co.uk/service/

URL B

https://www.example.co.uk/service/?source=newsletter

If URL B contains:

<link rel="canonical" href="https://www.example.co.uk/service/" />

You are indicating that URL A is your preferred version.

Search engines can then focus their indexing and ranking signals on the preferred URL rather than treating every variation as an entirely independent page.

Other signals can also influence canonicalisation, including:

• Permanent redirects
• Internal links
• XML sitemap URLs
• HTTPS rather than HTTP
• The actual similarity of the page content
• Language and regional annotations
• Technical accessibility of each URL

Official search documentation describes permanent redirects and rel="canonical" annotations as strong canonicalisation signals, while sitemap inclusion is a weaker signal. Using consistent signals makes your preference clearer.

Why Are Canonical Tags Important for SEO?

Canonical tags help solve several technical SEO problems at once.

They consolidate signals between duplicate URLs

If links point to several versions of effectively the same page, canonicalisation can help search engines consolidate those signals around the preferred URL rather than treating each variation separately.

This is particularly useful when tracking parameters, filters or other URL variations have attracted links of their own.

They make the preferred search result clearer

You may not want an untidy parameter URL appearing in search results when a short, descriptive version of the page exists.

A well-configured canonical helps indicate which URL should normally represent the content.

They can simplify performance reporting

If the same content exists under several URLs, traffic and search performance can become fragmented across those URLs.

Establishing a consistent preferred version makes reporting easier to interpret.

They can reduce unnecessary crawling of duplicates

Search engines have limited crawling resources. On a large website, hundreds or thousands of parameter combinations can create unnecessary URLs.

Canonicalisation can help search engines understand that those URLs belong to the same content cluster, although it should not be treated as a complete crawl-management system.

They help maintain a cleaner website structure

Canonicalisation works best alongside good internal linking.

All of your important internal links should normally point directly to the canonical URL rather than unnecessarily linking through duplicates.

This is similar to the wider structural issue created by orphan pages: a technically valid page can still perform poorly when the rest of the website sends unclear structural signals.

When Should You Use a Canonical Tag?

A canonical tag is most useful when multiple accessible URLs need to remain available but one should be treated as the preferred version.

URL Tracking Parameters

Marketing and analytics parameters can create URLs such as:

/seo-guide/?utm_source=newsletter

/seo-guide/?campaign=october

The content has not materially changed. The parameters merely provide tracking information.

Both parameter URLs can therefore point canonically to:

/seo-guide/

Ecommerce Filtering and Sorting

An ecommerce category might be accessible as:

/chairs/

/chairs/?sort=price

/chairs/?colour=blue&sort=price

The correct approach depends on whether the filtered page has independent search value.

If a filtered URL simply rearranges substantially the same products, canonicalising it to the main category may be appropriate.

If the filtered page serves a genuinely different search intent and you deliberately want it indexed, automatically canonicalising it back to the parent category could prevent that URL from being treated as a standalone search page.

This is why blanket canonical rules on ecommerce sites need careful planning.

Product Variants

Products may generate separate URLs for colour, size or configuration options.

If every variation presents essentially the same product information, one primary product URL may be the logical canonical.

However, if a variation has distinct content, demand, availability or a meaningful search intent of its own, it may deserve its own indexable URL.

Print and Alternative Page Versions

A printer-friendly version may display almost exactly the same information as the standard webpage.

In that situation, the print version can point canonically to the normal webpage while remaining available to users.

Tracking and Session URLs

Some systems append session IDs or other temporary parameters.

Where those variations do not alter the main content, canonicalisation can stop them being treated as independent versions.

Content Available in Several Technical Locations

Website migrations, staging systems, content-management changes and development work can occasionally leave similar content accessible under more than one route.

These situations should be investigated rather than automatically solved with canonical tags.

If one version should no longer exist at all, a permanent redirect will often be the cleaner solution.

For websites built using WordPress, correct canonical configuration should form part of wider WordPress SEO rather than being treated as an isolated setting.

Canonical Tags vs Redirects vs Noindex

Canonical tags, redirects and noindex directives solve different problems.

Method	Main Purpose	Can User Visit Original URL?	Should Original URL Remain Indexed?
Canonical	Identify the preferred version of similar content	Yes	Usually not as the preferred version
Permanent redirect	Move visitors and search engines to another URL	No	Normally replaced by destination
Noindex	Prevent a page appearing in search results	Yes	No
Sitemap	Indicate URLs you consider important	Yes	Potentially

Use a canonical when several versions genuinely need to remain accessible.

Use a permanent redirect when an old or duplicate URL no longer needs to exist independently.

Use noindex when the page can remain accessible to visitors but should not appear in search results.

Do not use noindex as a substitute for canonicalisation when your real objective is to consolidate two versions of the same content. Official guidance specifically recommends rel="canonical" rather than noindex for that purpose.

When Should You Not Use a Canonical Tag?

Canonical tags are powerful, but they are frequently used to hide structural problems that should be fixed another way.

When the old page should disappear

If Page A has permanently moved to Page B, redirect Page A.

Leaving both pages accessible and merely adding a canonical creates an unnecessary duplicate.

When the pages are genuinely different

Do not canonicalise two pages just because their topics are related.

For example, these should normally remain separate:

/wordpress-seo/

/local-seo/

They address different requirements even though both belong to SEO.

A canonical should not be used as an attempt to transfer authority between unrelated pages.

When you simply do not want a page indexed

If a page has a genuine reason to remain outside search results, a suitable indexing directive may be required instead.

Canonicalisation does not guarantee exclusion from an index.

When pages target separate search intentions

A category page, product page, service page and informational guide may contain overlapping terminology while serving completely different purposes.

Similarity alone does not make one a duplicate of another.

How to Add a Canonical Tag

For a standard HTML page, the canonical element belongs inside the <head> section.

A simplified example is:

<head>

<title>Example Page</title>

<link rel="canonical" href="https://www.example.co.uk/example-page/" />

</head>

The URL should normally be absolute rather than relative.

Use:

https://www.example.co.uk/example-page/

rather than:

/example-page/

Absolute URLs reduce ambiguity and are specifically recommended in current search documentation.

Should a page canonicalise to itself?

In most normal situations, yes.

A self-referencing canonical means the preferred page contains a canonical tag pointing to its own URL:

<link rel="canonical" href="https://www.example.co.uk/example-page/" />

This reinforces which URL you regard as the correct version and can protect against accidental duplicates created by parameters or alternative routes.

Self-referencing canonicals are recommended as a canonicalisation best practice.

What about non-HTML files?

Canonicalisation is not limited entirely to HTML pages.

For documents such as PDFs, a canonical relationship can be communicated through an HTTP response header instead of inserting an HTML element into the document itself.

How to Check a Canonical Tag

There are several straightforward ways to inspect canonicalisation.

View the HTML source

Open the source code for the page and search for:

rel="canonical"

Confirm:

• A canonical exists where expected
• Only one intended canonical is present
• The URL is correct
• HTTPS is being used
• The target page is accessible
• The URL does not contain an accidental parameter
• The canonical appears within the <head>

Crawl the website

A technical website crawl can identify:

• Missing canonicals
• Multiple canonical elements
• Broken canonical destinations
• Canonicals pointing to redirected URLs
• Non-indexable canonical destinations
• Canonical chains
• Pages canonicalised to unrelated content

This is particularly useful on ecommerce websites and larger sites where manually inspecting every URL is impractical.

Compare canonical URLs with your sitemap

The URLs you place in your XML sitemap should normally be the URLs you want indexed.

If your sitemap contains:

/page-a/

but /page-a/ canonically points to /page-b/, you are sending conflicting signals.

Review your internal links

Your navigation, breadcrumbs, category links and contextual links should normally point directly to canonical URLs.

Official guidance specifically recommends linking internally to the canonical version rather than its duplicate.

Common Canonical Tag Mistakes

Small canonical errors can affect large numbers of pages when they are generated automatically by a content management system.

Using more than one canonical

A page should not deliberately provide competing canonical destinations.

For example:

Canonical → /page-a/

and

Canonical → /page-b/

creates ambiguity.

Choose one preferred URL and make the technical signals consistent.

Pointing canonicals at broken pages

A canonical destination should be a valid page.

Pointing towards a 404, removed URL or otherwise inaccessible destination undermines the purpose of identifying a preferred page.

Canonicalising to a redirected URL

Avoid:

Page A → canonical → Page B → redirect → Page C

Where Page C is genuinely the preferred destination, Page A should normally reference Page C directly.

Canonicalising HTTPS pages to HTTP

Where the secure HTTPS version is the intended URL, the canonical should normally reference HTTPS as well.

Search documentation states a preference for HTTPS over equivalent HTTP pages unless conflicting technical signals exist.

Putting the canonical outside the head

For an HTML canonical link element to be interpreted correctly, it should appear in the <head> section of valid HTML.

Using relative URLs unnecessarily

Although a relative canonical can technically work, an absolute URL is clearer and reduces the chance of errors when content is moved or copied between environments.

Sending conflicting signals

A common configuration looks like this:

• Sitemap lists Page A
• Internal links point to Page A
• Canonical points to Page B
• Redirects point somewhere else

The cleaner your signals, the easier it is for a search engine to understand your intention.

Automatically canonicalising everything to the homepage

This does not make the homepage stronger.

It simply tells search engines that large numbers of unrelated pages supposedly duplicate the homepage, which they do not.

Canonical relationships should be based on duplicate or very similar content.

Canonicalising every paginated page to page one

Pages two, three and four of a sequence are not necessarily duplicates of page one.

They contain different items and may need to remain independently crawlable.

Do not use a site-wide rule that blindly points every paginated URL back to the first page.

Why Might a Search Engine Ignore Your Canonical Tag?

Because canonical tags are signals rather than commands, a search engine may choose another URL.

Common reasons include:

• The two pages are not similar enough
• Internal links favour a different URL
• The sitemap lists another version
• The chosen canonical redirects elsewhere
• The canonical destination returns an error
• Several canonical tags conflict
• HTTP and HTTPS signals conflict
• JavaScript alters the canonical after the original HTML loads
• Different localisation signals point elsewhere

The correct response is not usually to add more canonical tags.

Look at every signal pointing towards the affected URLs and make them consistent.

Canonical Tags and JavaScript

JavaScript can complicate canonicalisation if the canonical tag changes after the initial HTML has been delivered.

Where possible, make the preferred canonical clear in the original HTML.

Current search-engine guidance recommends avoiding situations where JavaScript changes a canonical that was already declared differently in the original source.

For JavaScript-heavy websites, always compare:

• Raw HTML
• Rendered HTML
• Canonical destination
• Internal links
• Sitemap URLs

A crawler viewing only the final rendered page may not reveal the full sequence of events.

Canonical Tags on Ecommerce Websites

Ecommerce websites are particularly prone to duplicate URL generation.

One category can create variations for:

• Price
• Colour
• Size
• Manufacturer
• Availability
• Rating
• Sorting order
• Number of products displayed
• Tracking parameters

Before adding a canonical rule, decide whether each filtered combination deserves to exist in search.

For example, a filtered category for a product attribute with meaningful search demand may be worth optimising as an independent landing page.

A sort-by-price version containing exactly the same products in another order probably has far less reason to be indexed independently.

The important question is therefore not:

“Does this URL contain a parameter?”

It is:

“Does this URL represent genuinely useful, independently valuable content?”

That distinction prevents canonicalisation from accidentally removing useful landing pages from your organic search strategy.

Canonical Tags and Location Pages

Location pages should not be canonicalised together merely because they use a similar layout.

A well-built location page should contain information specific to the place and service it covers.

If pages for several towns contain genuinely different local information and are intended to rank separately, each would normally use its own self-referencing canonical.

Pointing every location page back to one national page can undermine the signal that those individual pages are intended to stand independently.

The better solution is to make each location page useful enough to justify its existence.

This principle also applies when developing a local SEO strategy: canonical tags cannot compensate for thin, cloned or poorly differentiated local pages.

How Do Canonical Tags Affect Internal Linking?

Canonicalisation and internal linking should reinforce one another.

Suppose these two URLs show the same content:

/technical-seo/

/technical-seo/?ref=menu

If /technical-seo/ is canonical, your permanent navigation and contextual links should ideally point directly to /technical-seo/.

Consistently linking to the parameter version creates unnecessary crawling and sends mixed structural signals.

You should therefore review canonical URLs alongside:

• Navigation
• Breadcrumbs
• Related-content links
• XML sitemaps
• Product-category links
• HTML sitemaps
• Pagination
• Redirect rules

Canonicalisation is strongest when the rest of the website agrees with it.

Should Canonical Pages Appear in the XML Sitemap?

Generally, your sitemap should contain the URLs you want treated as the primary indexable versions.

Avoid filling it with tracking URLs, redirected URLs or duplicates that immediately canonicalise somewhere else.

Sitemap inclusion is considered a weaker canonical signal than redirects or rel="canonical", but consistency between all three signals can strengthen the overall indication of which URL you prefer.

A Practical Canonical Tag Checklist

Before publishing or changing canonical tags, check the following:

• Is the chosen URL genuinely the preferred version?
• Is the canonical destination substantially the same content?
• Does the preferred URL return a successful response?
• Is the destination indexable?
• Does it use HTTPS?
• Is the URL absolute?
• Is there only one intended canonical?
• Does the canonical appear inside the HTML head?
• Does the preferred page reference itself?
• Do internal links use the preferred URL?
• Does the XML sitemap contain the preferred URL?
• Are redirects consistent with your choice?
• Are regional or language versions configured appropriately?
• Does JavaScript leave the canonical unchanged?
• Have you avoided canonicalising genuinely unique pages?

For larger sites, these checks should be included in regular technical SEO audits rather than carried out only after rankings or indexation problems appear.

Frequently Asked Questions

Does every webpage need a canonical tag?

Not every webpage technically requires an explicitly declared canonical, but self-referencing canonicals are a useful best practice on normal indexable pages because they clearly indicate your preferred URL.

Can a canonical tag stop duplicate content?

It helps search engines understand which version should represent a set of duplicate or very similar pages. It does not delete the other URLs or physically prevent users from visiting them.

Is a canonical tag the same as a redirect?

No. A canonical leaves both URLs accessible to users. A redirect automatically moves the visitor from one URL to another.

Can I canonicalise a page that has completely different content?

You should not. Canonical tags are designed for duplicate or substantially similar content rather than unrelated pages.

Can I use noindex and canonical together?

Mixing them can create unclear signals because one asks for exclusion while the other identifies another URL as the representative page. If your objective is canonicalisation, use the canonical approach consistently rather than adding noindex as a second solution.

How long does a canonical change take to be recognised?

There is no fixed timescale. The affected URLs need to be crawled and processed again before a new canonical relationship can be reassessed. Larger or frequently crawled websites may see changes recognised sooner than pages that are crawled infrequently.

Should tracking URLs have canonical tags?

Where a tracking parameter does not materially change the page content, it is usually sensible for that URL to reference the clean preferred version.

Should location pages canonicalise to the main service page?

Not if those location pages contain genuinely useful, unique local content and are intended to appear independently in search. Each should normally be treated as its own indexable page.

Final Thoughts

Canonical tags are simple pieces of code, but they sit at the centre of how duplicate and alternative URLs are organised.

The most reliable approach is to decide which URL should represent each piece of content and make the entire website support that choice.

Your canonical tags, redirects, internal links, XML sitemap, HTTPS configuration and page content should all tell the same story.

If they disagree, a search engine may make the decision for you.

Author note: Written by Terry Burrows of Search Focus, a freelance SEO consultant working in search optimisation since 2002.