A canonical tag is an HTML link element placed in the <head> that tells search engines which URL you want treated as the main (preferred) version when the same or very similar content exists at multiple URLs.
This matters because duplicate or near-duplicate URLs can split ranking signals, confuse reporting, and waste crawl time, which is why it’s closely tied to duplicate content in SEO.
Here’s what it looks like in practice: <link rel="canonical" href="https://www.example.com/your-page/">. We usually recommend a self-referencing canonical on indexable pages (the page points to itself) and then only change it when there’s a real duplicate situation.
When you should use a canonical tag
- Tracking parameters create extra URLs (UTMs, click IDs, session IDs) that show the same page, so you canonical the clean URL.
- Ecommerce filters and sort orders generate many versions of the same category or product listing, so you canonical to the main version you want shown in search.
- Multiple URL paths to the same product (for example, a product reachable from multiple categories) so one product URL is the preferred version.
- Print-friendly pages or alternate layouts that repeat the same content, so the main page stays the preferred one.
- Syndicated or republished content when you control both sides and want the original page to be treated as the main version, including between domains.
If you want us to audit and clean up canonical issues without turning it into a tech project, our SEO services usually start by mapping out your “clean” URLs and then removing the duplicates created by parameters, category paths, and platform quirks.
When not to use a canonical tag
Don’t use canonicals to “hide” pages you actually want ranking. If a page is meant to stand on its own (different service, different location, different intent), it needs its own unique content and a self-canonical. For example, an Orlando dental practice should not point every neighborhood or service page back to one generic page if those pages are meant to attract different searches.
| Situation | Use a canonical? | Better fix if you can |
|---|---|---|
| HTTP vs HTTPS or www vs non-www versions | Sometimes, but not ideal | Use a single preferred version with redirects |
| Page was moved or removed | No | Use a 301 redirect to the new page |
| Pagination (page=2, page=3) | Yes, but each page should point to itself | Clean internal linking and index rules for filters |
| Filtered/sorted URLs you do not want in search | Maybe | Use noindex for pages you don’t want indexed |
Canonicals are a strong signal, but they’re still a hint, so search engines can pick a different preferred URL if other signals disagree (redirects, internal links, sitemap URLs, or inconsistent canonicals). If you’re seeing “Google chose a different canonical” in Search Console, it usually means your site is sending mixed signals that need cleanup.
For most businesses, the cleanest setup is: one preferred URL per page, one canonical per page, and redirects for true duplicates like old URLs or alternate site versions, which is why 301 vs 302 redirects matters here.
If you’re on WordPress or a builder and you’re not sure where the canonical is coming from, it’s often tied to theme and plugin settings, URL settings, and how pages are generated, so our web design team typically checks canonicals as part of technical SEO hygiene during rebuilds and redesigns.
