What a canonical tag actually does
Ecommerce platforms routinely serve the same product or category through multiple URLs - with and without a trailing slash, with different sorting or filter query parameters, or reachable from more than one category path. Without a canonical tag, search engines have to guess which version to index, which can split ranking signal across near-duplicate URLs instead of consolidating it on one.
The canonical tag sits in a page's <head> and points to the single URL that should be treated as the master copy. Every duplicate or near-duplicate variant carries a canonical tag pointing back to that one URL; the master URL typically has a canonical tag pointing to itself.
Why it matters during a platform migration
New platforms often introduce their own URL patterns for filtering, pagination, or sorting that didn't exist on the old platform, and it's easy to end up with unintended duplicate content if canonical tags aren't set correctly from day one. Getting this wrong doesn't usually cause a dramatic drop like a missing redirect does, but it quietly dilutes ranking signal across near-duplicate pages instead of consolidating it - a slower, harder-to-diagnose problem.
Migrations are also a natural point to fix pre-existing canonicalization issues from the old platform, rather than carrying them forward unchanged.
Canonical tags vs. redirects
A canonical tag and a redirect solve different problems. A redirect (see 301 redirect) is for a URL that no longer exists at all - it sends visitors and crawlers to the new address. A canonical tag is for URLs that still exist and are still reachable, but that shouldn't compete with each other in search results - both stay live, one is just marked as the reference version.
Frequently asked questions
Does a canonical tag redirect visitors to the main URL?
No. Visitors can still land on the non-canonical URL and it loads normally - the canonical tag only signals to search engines which version to index and rank, it doesn't change what a browser does.
Do I need canonical tags if I don't have duplicate content?
It's still common practice for every page to have a self-referencing canonical tag (pointing to itself), which removes ambiguity if a duplicate ever does appear later - e.g. through a tracking parameter or a new filter URL.
Can a canonical tag point to a page on a different domain?
Yes, cross-domain canonicals are valid and are sometimes used deliberately, but for a platform migration they're usually not what you want - each new URL should typically canonicalize to itself, with redirects (not canonicals) handling the old domain or old URLs.
What happens if my canonical tags point to the wrong URL?
Search engines may end up indexing and ranking the wrong version of a page, or ignoring your canonical signal altogether if it looks inconsistent with other signals (like internal links or the sitemap). We check canonical tags as part of migration QA rather than assuming the new platform sets them correctly by default.