canonical URL

A canonical URL is the address a site formally names as the “real” version of a page, used whenever more than one web address could plausibly show the same content. A product page reachable both with and without a tracking parameter, or under both the www and non-www version of a domain, is technically two or more addresses showing one page. The canonical URL tells search engines which one to treat as the real one and fold the others into.

Without it, search engines have to guess, and their guess is not always the one you would have picked. Worse, ranking signals like links can get split between the versions instead of combining behind the one address that should actually rank.

Why it matters to you

Search engines have a limited appetite for treating near-identical pages as separate, valuable things. A site with the same content living under several slightly different addresses, common after a redesign, a platform migration, or URL parameters added by filters and tracking, risks competing against its own duplicates instead of one strong page winning clearly.

It also matters for anything you actually promote. If you share a link with a tracking parameter attached and a search engine treats that as a distinct page from the clean version already ranking, you can accidentally undercut your own indexing rather than helping it.

How I set it up

On the sites I build, every page declares its own canonical address automatically as part of how the page is generated, so there is never a manual step where someone has to remember to set it correctly. If the same content is genuinely reachable through more than one path, which happens less on a modern, code-based build than on older platforms, the canonical tag points every version back to the one address that should count.

I check this whenever I inspect a site’s technical setup, because it is one of the more common problems inherited from platform migrations, where old URL structures and new ones end up both live and competing without anyone intending it.

What it looks like in practice

You will not see this directly unless you look at a page’s source code, where a single line points to the canonical address. What you will see, indirectly, is a site where one clear version of each page accumulates the inbound links and ranking signal, instead of it being split thin across duplicates nobody chose to create.

It is quiet, technical housekeeping. Done right, it never comes up. Done wrong, it can quietly cap how well a genuinely good page performs, for a reason that has nothing to do with the content itself.