What a canonical tag does
A website can make the same page available at more than one URL. Tracking parameters, sorting options, alternate paths and variations in URL formatting can all create duplicates or near-duplicates. A canonical tag helps you tell search engines which URL you would prefer them to treat as the main version. On an HTML page, it is placed in the document head as a link element with rel="canonical" and an href pointing to the preferred URL.
A canonical tag is a signal, not a command. A search engine may select a different canonical URL if other signals suggest that another version is more suitable. The tag also does not move visitors or prevent them from opening the page where it appears. Its purpose is to clarify the relationship between similar URLs, not to replace clear navigation or fix every indexing problem.
Choose a URL that can serve as the main version
Start by identifying pages that present substantially the same content. For each group, choose a preferred URL that works reliably and represents the content you want people to find. It should be accessible, return a successful response, and contain the content you expect search engines to associate with the group. A URL that redirects, returns an error, or leads to an unrelated page is a poor canonical destination.
Use the full preferred URL in the tag, and keep the choice consistent across the site. Internal links, sitemap entries and redirects should not repeatedly point to competing versions if you have chosen one as primary. A self-referencing canonical tag on the preferred page can make that choice explicit, particularly when optional parameters may be appended to its URL.
Do not group pages merely because they share a template or a few paragraphs. Product variants, location pages and other similar-looking pages may each offer distinct information. If a visitor would reasonably need to see both pages, pointing one canonical tag at the other may obscure useful content rather than resolve duplication.
Know when another approach fits better
Use a redirect when an alternate URL no longer needs to be available to visitors. A redirect takes users and crawlers to the destination; a canonical tag leaves the original URL accessible. For example, if an old page has been permanently replaced, redirecting it is usually clearer than serving the old page with a canonical tag.
A canonical tag is also different from a noindex directive. Noindex asks search engines not to show a page in results, while canonicalization expresses a preference about which version of similar content should be treated as primary. Combining them on the same page can make your intent harder to interpret. Likewise, blocking a duplicate URL from crawling can prevent a crawler from seeing the canonical tag placed on it.
Keep paginated pages in mind: later pages in a series often contain items that the first page does not. Pointing every page in the series to page one can misrepresent those differences. Assess whether each page should remain discoverable before deciding how to handle it.
Implement the tag without creating conflicts
Place one clear canonical link element in the head of each HTML page that needs one. A page should not declare several different preferred URLs. Check that the tag is present in the version of the page search engines can receive, especially when a site generates head elements with JavaScript or applies different templates to different sections.
Avoid canonical chains in which one page points to a second page that points to a third. Point duplicate pages directly to the chosen URL. Also avoid circular signals, such as two pages naming each other as canonical. If duplicate content spans separate domains you control, a cross-domain canonical can express a preference, but it should be used only when the pages genuinely correspond and you intend one version to be primary.
For files that do not have an HTML head, some systems can provide a canonical signal through an HTTP response header. This is a more specialized setup; confirm that the server sends the intended header on the relevant file responses rather than assuming an HTML tag will cover them.
Check the result and maintain it
After publishing, inspect both the duplicate page and the preferred page. Confirm that the duplicate declares the intended destination, the destination loads correctly, and the pages remain similar enough to justify the relationship. Check a sample from each page template, not just a single URL: a template error can affect an entire section.
If you use a search engine’s indexing reports, compare your declared canonical with the URL it selected, where that information is available. A different selection is a reason to investigate conflicting links, redirects, content or tags, not proof that the tag is broken. Recheck canonical signals when you change URL structures, templates or navigation. The goal is a consistent, understandable path to the version of each page that best serves search users.
