Canonical Tag Checker: Fix duplicate content.

What a canonical tag actually does

Canonical tags tell Google which URL is the “real” version of a page when several URLs render similar or identical content. They are easy to get wrong, and a wrong canonical can quietly send ranking signals to the wrong URL — or to a URL Google won’t index at all.

Best practice is a self-referencing canonical: the page names itself, as an absolute HTTPS URL that matches your preferred host (www or not) and your trailing-slash convention. Pointing at an older version, a different protocol, the homepage, or a 404 tells Google to consolidate signals somewhere other than where you intended.

Canonical mistakes to look for

Our scan covers the first item below on 30 pages sampled evenly across your sitemap. Use the rest of the list as a manual checklist.

• Canonical points elsewhere — the page names a different URL as canonical, so Google may index that URL instead. Our scan flags this.
• Canonical chains — A points to B, and B points to C. Google treats canonicals as hints, and a chain weakens the hint.
• Protocol mismatches — an HTTP page with an HTTPS canonical, or the reverse; the version you didn’t want may be the one indexed.
• Host mismatches — a bare-host page declaring a www canonical, or the reverse, when the rest of the site is consistent. It splits signals between two hosts.
• Trailing slash inconsistency — a page at /about with canonical /about/, or the reverse. A tiny detail that can split signals between two URLs.
• Cross-page canonicals — a product variant pointing its canonical at the parent product. Sometimes correct, sometimes a bug; our scan flags it so a human can decide.
• Multiple canonical tags — two or three rel=canonical tags on one page; Google may ignore all of them.
• Canonical to a 404 — pointing at a URL that returns 404 or 410. Google can’t use that hint, so it picks a canonical itself.

Why canonical bugs spread fast

Canonical tags are usually rendered by a CMS template, which means a single template bug affects thousands of pages at once. WordPress sites typically inherit canonicals from Yoast/Rank Math; misconfigured plugins generate canonicals to the wrong protocol after a TLS migration. Headless sites built in Next.js or Astro often hardcode the canonical domain, which breaks when the site moves to a new domain or adds a staging environment. E-commerce platforms (Shopify, WooCommerce) use parametrised URLs (?variant=X) that need their own canonical handling, and the defaults are often wrong out of the box.

Our scan samples 30 URLs from your sitemap, reads the canonical tag in each page’s HTML, and compares it with the page’s own URL. It ignores http/https and trailing-slash differences, so it flags only canonicals that point to a different host or path. The AI action plan then ranks what to fix first. Read our deeper coverage on the canonical tags learn page for the full taxonomy of canonical mistakes.