A canonical tag tells search engines which version of a page is the one you want indexed when several addresses show substantially the same content. It is a strong hint rather than a command, and it is not a fix for pages that should not exist. Most small sites need it in three specific situations and nowhere else.
What it does
Where the same content is reachable at several addresses, a search engine must pick one to represent them. The canonical tag is your statement of preference.
Two things follow from that, and both are routinely misunderstood.
It is a hint, not an instruction. Google considers your declared canonical alongside other signals, including internal links, redirects, and sitemaps, and can choose differently. Search Console reports where it has, which is useful information about whether your signals agree with each other.
It does not remove a page. The other versions remain accessible. If a page genuinely should not exist, the answer is a redirect or removal, not a canonical pointing elsewhere.
The three situations that actually apply
1. Parameters on the end of URLs
Tracking parameters, session identifiers, and filter values all create distinct addresses for the same page. A canonical pointing at the clean version consolidates them.
This is the most common genuine need on a small site, and it usually arrives via advertising or email campaign links.
2. Address variations
With and without www, with and without a trailing slash, http and https. Redirects are the better tool here, and a self-referencing canonical on every page provides a useful backstop while ensuring each page declares its own preferred form.
3. The same content in more than one place
A product appearing in two categories, an article listed under two topics, or a printable version alongside a standard one. Where both must exist, canonical tags nominate the primary.
Self-referencing canonicals
Every page carrying a canonical tag pointing at itself is normal practice and worth having. It costs nothing, it removes ambiguity when a page is reached with parameters attached, and it is what most content management systems do by default.
It is not a magic improvement, and anyone selling it as one is overstating it. It is hygiene.
The mistakes
- Every page canonicalised to the homepage. A surprisingly frequent misconfiguration, and it asks search engines to ignore your entire site except one page. Usually a plugin setting applied without understanding.
- Canonical pointing at a redirecting page. Contradictory signals: the tag says this is the primary version while the destination says it has moved.
- Canonical pointing at a blocked or missing page. Same problem, more severe.
- Using it instead of a redirect. If the old page should not exist, redirect it. A canonical leaves it accessible and hopes nobody notices.
- Multiple canonical tags on one page. Conflicting declarations, usually from a theme and a plugin both adding one. The result is that both may be ignored.
- Relative rather than absolute addresses. Use the full address including protocol and domain.
When you do not need one
Being clear about this, because canonical tags get applied defensively where nothing requires them.
- Pages with genuinely different content. Two service pages covering different services need no relationship declared.
- Pages that simply resemble each other. Similar is not duplicate, and the concern is misplaced, as covered in duplicate content.
- Pages you have redirected. The redirect has already settled it.
How to check
View the source of a few pages and search for the canonical declaration. Confirm it points where you expect, uses the full address, and appears once.
Then look at Search Console, which reports the canonical Google selected against the one you declared. A disagreement is not automatically a problem, and it is a reliable indicator that your internal links, sitemap, and tags are not all saying the same thing.
Frequently asked questions
What does a canonical tag do?
It tells search engines which version of a page you want indexed when substantially the same content is reachable at several addresses. It is a strong hint considered alongside other signals rather than a command, and search engines can select a different version.
Does a canonical tag remove a page from a site?
No. Other versions remain fully accessible to anyone with the address. Where a page genuinely should not exist, the appropriate tool is a redirect or removal rather than a canonical tag pointing somewhere else.
Should every page have a self-referencing canonical?
It is normal practice and worth having. It costs nothing, removes ambiguity when a page is reached with tracking parameters attached, and most content management systems add it automatically. It is hygiene rather than an improvement.
What is the most common canonical tag mistake?
Every page on a site canonicalised to the homepage, usually from a plugin setting applied without understanding what it does. The effect is asking search engines to disregard the entire site apart from one page.
When do I not need a canonical tag?
Where pages have genuinely different content, where pages merely resemble one another without being duplicates, and where you have already implemented a redirect, since the redirect has settled the question already.
How do I check my canonical tags are correct?
View the page source and confirm the canonical declaration points where you expect, uses the full address including protocol and domain, and appears only once. Search Console also reports where Google selected a different canonical than the one declared.
West Coast Media Solutions Inc. builds and maintains websites for businesses across Canada. Technical behaviour described here reflects how search engines and browsers currently work; specifics change, and your own configuration is what determines the outcome on your site.
Something technical not behaving?
We build and maintain websites for businesses across Canada, and we sort out this kind of thing without the drama.
Start a Conversation