Canonical URL (Canonical URL) is the URL that a search engine accepts as the “primary/representative” version of a web page. In cases where the same or very similar content exists at multiple URLs, it is specified with the <link rel="canonical" href="..."> tag.
Common cases where canonical should be used:
- HTTP and HTTPS versions existing at the same time
- WWW and non-WWW versions
- Sorting, filtering, and tracking parameters in the URL (utm_*, ?sort=)
- Serving mobile and desktop on separate URLs
- The same product being listed under different categories
- Printable versions (?print=1)
- Paginated listing pages
The canonical tag is not a directive, but a hint. Google makes its own choice when there are conflicting signals and may designate a different URL as canonical. This is reported in Search Console as “Google-selected canonical.”
Tip: Every page should include its own self-referencing canonical (that is, the page indicates its own URL as canonical). This minimizes the risk of misinterpretation and provides insurance against duplicate content issues.