Canonical tags do not fix duplicate content: they only suggest a preferred URL

Canonical tags do not fix duplicate content: they only suggest a preferred URL

A canonical is a hint, not a command. Search engines may ignore it when signals conflict: internal links point elsewhere, the page is thin, the duplicate has stronger backlinks, or the URL is blocked by robots. If the rest of the site says “index this page,” the canonical alone rarely wins.

For faceted navigation, treat duplicates by type:
— Same page, different tracking parameters: strip or normalize parameters.
— Sorts, filters, pagination: decide whether each URL deserves indexation.
— Near-duplicates with unique intent: add content, not just a canonical.
— Junk parameter combinations: prevent crawl traps with rules, not hope.

A clean canonical only works when everything else agrees: self-referential canonicals on indexable pages, consistent internal links, one URL per template, and no mixed signals from sitemap or hreflang. If the page should not rank, use noindex or block crawling carefully; if it should rank, make it genuinely distinct.

Use canonicals as part of a duplicate strategy, not as the strategy itself.