TL;DR
- Duplicate URLs usually come from tracking parameters, http and https or www versions, sorting options and printer-friendly pages.
- rel=canonical is a strong hint, not a directive; Google may choose a different canonical if other signals disagree.
- Redirects, internal links, sitemaps and canonical tags should all point to the same URL.
- There’s no penalty for ordinary duplicate content; the risk is diluted signals and the wrong URL showing in results.
Search engines think in URLs. If the same content is available at https://example.com/services, https://www.example.com/services/ and https://example.com/services?utm_source=newsletter, that’s three URLs competing to represent one page.
Where duplicates come from
| Source | Example |
|---|---|
| Protocol and host | http vs. https; www vs. no www |
| Trailing slashes and case | /services vs. /services/ vs. /Services |
| Tracking parameters | ?utm_source=…, ?fbclid=… |
| Sorting and filtering | /shoes?sort=price, /shoes?color=red |
| Session IDs and print versions | ?sessionid=…, /print/ |
| Syndication | Your article republished on another site |
How canonical tags work
Add a link element in the page’s head pointing to the preferred URL:
<link rel="canonical" href="https://www.example.com/services/">
Google treats this as a strong hint. It weighs it alongside other signals, including redirects, the URLs in your sitemap and the URLs your internal links use. If those signals contradict your canonical tag, Google may pick a different URL, and the page indexing report will show “Google chose different canonical than user.”
Make the signals agree
- Pick one host and protocol, such as https and www, and 301-redirect the others to it.
- Use absolute canonical URLs that match that choice exactly.
- Link internally to the canonical URL, not to parameter or alternate versions.
- List only canonical URLs in your sitemap; see XML sitemaps.
- Don’t canonicalize page 2 of a list to page 1; paginated pages show different content.
Common mistake: A common template bug sets every page’s canonical to the homepage. Google then has conflicting signals across your whole site. Spot-check canonical tags on a few different page types after any theme or plugin change.
Is duplicate content a penalty?
Not in the ordinary sense. Google’s guidance is that duplicate content on a site isn’t grounds for action unless it’s meant to deceive. The real cost is practical: links and signals split across URLs, and the wrong version showing up in results.
FAQ
Should every page have a self-referencing canonical tag?
It’s a good habit. A canonical tag pointing to the page’s own clean URL helps when the page is reached with tracking parameters or other variations.
Can I canonicalize to a different domain?
Yes, cross-domain canonicals are supported, for example when content is syndicated. Google still treats them as a hint.
Is a canonical tag better than a redirect?
If users don’t need the duplicate URL at all, a 301 redirect is the stronger signal. Use a canonical when the duplicate must stay accessible, such as a filtered product list.
Search engines update their systems and documentation often. This guide reflects Google’s published guidance at the time of writing; check Google Search Central for current details, and test changes on a staging copy of your site before making them live.