A canonical URL tells search engines which version of a page is the original. When the same content is accessible at multiple URLs, canonical tags prevent those versions from competing against each other and diluting your ranking signal.
Common scenarios where you need a canonical:
- Your page is accessible with and without trailing slash (example.com/page and example.com/page/)
- HTTP and HTTPS versions both exist
- www and non-www versions both resolve
- Filtered or sorted versions of a page (example.com/products?sort=price)
- Paginated content where page 1 also has all the content
How to implement it:
<link rel="canonical" href="https://example.com/the-original-page">
The canonical should always use the full absolute URL including the protocol (https://). Never use a relative URL in a canonical tag.
Self-referencing canonicals are canonical tags that point to the current page itself. These are good practice even when you do not have a duplication problem — they explicitly declare intent and prevent future issues if the page becomes accessible at a different URL.
Cross-domain canonicals tell Google that content on your site is a copy of content on another domain. Use these carefully — they pass ranking signals to the other domain.
The most common canonical mistake is setting no canonical at all. The second most common is setting different canonicals on different versions of the same page, which confuses Google rather than helping it.