A canonical tag is a line of code (<link rel="canonical" href="...">) placed in a page’s head that names the preferred URL for a piece of content. When the same or very similar content is reachable at several URLs, the canonical tag tells search engines which one to index and rank.
Why it matters
Duplicate URLs are common and often unintentional: tracking parameters, session IDs, HTTP and HTTPS variants, trailing slashes, and printer-friendly pages all create copies. Left unmanaged, they split ranking signals across versions and waste crawl resources.
A correct canonical consolidates those signals onto one URL, so the authority earned by every variant accrues to a single page rather than being divided.
How to use it
Point each duplicate or parameterized version at the clean, preferred URL. Self-reference the canonical on the primary page itself, so there is no ambiguity. Keep the canonical URL consistent with what appears in your sitemap and internal links.
A few rules keep canonicals working as intended:
- Use absolute URLs, not relative paths.
- Only one canonical per page.
- Do not canonicalize every page to the homepage, a frequent and damaging error.
A signal, not a directive
Search engines treat the canonical tag as a strong hint rather than an absolute command. If your internal links, sitemap, and canonical disagree, engines may pick a different URL than you intended. Consistency across all of those signals is what makes the canonical reliable. For the structured-data layer that complements clean canonicalization, see schema markup.