Canonical tag: how do you point Google to the right version of a page?
One page can answer at sixteen different addresses. The canonical tag says which one counts.
A canonical tag is a link rel=canonical element in a page's head that names the URL you want shown in search results when several URLs serve the same or very similar content. Google treats it as a strong signal rather than a command, so it can still pick a different version.
In short
- The canonical link relation is described in RFC 6596, published in April 2012.
- It can also be sent as an HTTP header, which is handy for PDFs.
- Redirects and canonical tags are strong signals; sitemap inclusion is weak.
- Google may pick a different canonical than the one you declared.
What is a canonical URL?
Google calls the canonical URL the one it should show in search when several pages have duplicate or very similar content [1]. To the search engine those pages form a cluster, and one of them represents the group.
The canonical link relation was written up by the IETF in RFC 6596, an informational document from April 2012. It describes canonical, informally, as the author's preferred version of a resource. Common uses it lists are URLs with added parameters such as session IDs, and preferring a single-page version over the same content split across several pages [2].
How can one page end up with many URLs?
Which methods send a canonical signal?
| Method | Signal strength | When to use it |
|---|---|---|
| Redirect | Strong | When you are retiring a duplicate URL |
| link rel=canonical in the head | Strong | HTML pages reachable at several URLs |
| rel=canonical HTTP header | Strong | Non-HTML files such as a PDF spec sheet |
| Listing in the sitemap | Weak | As a supporting hint for preferred URLs |
Source: Google Search Central [1].
Does Google always follow the canonical tag?
No. Declarations are signals, not orders. Without any, Google picks what it sees as the objectively best version to show. Even with a tag in place it weighs other signals, such as preferring HTTPS over HTTP and hreflang clusters [1].
When your signals disagree, say a tag pointing to A and an XML sitemap that only lists B, you are leaving the decision to Google.
What mistakes do people make with canonicals?
- Using robots.txt to handle duplicates: Google may still index the blocked URLs without seeing their content.
- Adding noindex to one version to choose a canonical within the same site, which removes that page from search entirely.
- Reaching for the URL removal tool, which hides every version of the URL.
- Declaring different canonicals for the same page through different methods, such as a tag and an HTTP header that disagree.
- Hard-coding one canonical into a page template so every page points to the homepage.
Should town pages point to one canonical?
A business serving Barrie, Orillia and Innisfil sometimes builds three near-identical pages. Pointing two of them at the third tells Google they are duplicates, and they may never show up. The better fix is to make each page genuinely useful for its town: response times, neighbourhoods served, local questions. See local SEO.
Does Pilotiko manage canonical tags?
Canonical management is not an advertised Pilotiko feature. The visibility features available for the site Pilotiko hosts are the sitemap, robots.txt, structured data, llms.txt and an audit of your current site. Those cover part of search engine optimization; canonicals are still yours to check.
Frequently asked questions
Should a page have a canonical pointing to itself?
Yes, Google recommends it: add rel=canonical to the canonical page itself, known as a self-referential canonical. It confirms the main address even when parameters get tacked onto the URL [1].
Canonical tag or redirect: which one?
If the old URL has no reason to exist, a redirect is the clearest signal. If both URLs need to stay reachable, for example a version with sorting parameters, use a canonical tag.
Can Google ignore my canonical tag?
Yes. Google treats it as a strong signal but can pick another URL it considers better, such as an HTTPS version over an HTTP one.
Can a PDF have a canonical?
Yes, by sending rel=canonical in an HTTP header rather than in HTML. That helps when a spec sheet exists both as a PDF and as a web page.
Should my English page point to the French one?
No. When you use hreflang, Google asks you to specify a canonical page in the same language. The English page points to itself and the two versions are tied together with hreflang [1].
Should internal links point to the canonical URL?
Yes. Google advises linking to the canonical URL rather than a duplicate, because consistent links help it understand your preference [1].
Sources
- Google Search Central, How to specify a canonical URL with rel="canonical" and other methods (accessed on October 15, 2026).
- IETF (RFC Editor), RFC 6596 : The Canonical Link Relation (accessed on October 15, 2026).
Read next
hreflang: how do you connect the English and French versions of a page?
hreflang links the English and French versions of a page so Google shows the right one. en-CA and fr-CA codes, return links, x-default and common mistakes.
XML sitemap: what does it do for search engines?
What an XML sitemap is, which tags it uses, the official 50,000-URL limit, whether a small business site needs one and the mistakes that confuse search engines.
robots.txt: what does this file tell search engine crawlers?
robots.txt tells crawlers which URLs they may visit. How RFC 9309 rules are read, what happens on errors, a small business example and the costly mistakes.
SEO: what is search engine optimization?
What SEO really covers, how Google finds and ranks pages, a worked example for a Quebec service business and the mistakes that quietly cost visibility.