Canonical URL
The one address you tell search engines to treat as the main version of a page that's reachable at several addresses.
A canonical URL is the address you tell search engines to treat as the main version of a page, when the same page can be reached at more than one address. You set it with a link tag in the page's HTML, <link rel="canonical" href="...">, or in an HTTP header.
Why pages end up with several addresses
It happens more often than you'd think. The same product page might be reachable as:
yourbrand.com/shoes/trail-runnerwww.yourbrand.com/shoes/trail-runneryourbrand.com/shoes/trail-runner?utm_source=newsletteryourbrand.com/sale/trail-runner
To a person, these are one page. To a crawler, they can look like four. Without a canonical URL, a search engine picks one itself, and it may not pick the one you want.
What a canonical does
It tells the search engine which address to show in results and which one gets the credit for links pointing at any of the others. It's a strong hint, not an order: if the page you point to is very different from the copy, Google may choose another. Google explains how it picks in its guide to consolidating duplicate URLs.
A redirect is stronger. If an old address should never be shown again, send it to the new one with a permanent redirect instead.
In AI answers
AI answers cite addresses too. When the same page lives at several, citations can be spread across them, which makes it harder to see how often that page is really used. One clear address per page, with a canonical on every copy, keeps the picture tidy.
In AnswerOrbit
Sources lists every page the answers used, by its address, and sorts your own pages from your site's exact address first. If you see the same page listed twice under different addresses, that's a sign it needs a canonical or a redirect. See Sources.