Duplicate URLs and canonical basics

To a person, example.com/shoes and example.com/shoes?utm_source=newsletter are the same page. To a search engine they are two different URLs that happen to return the same content. Most sites create duplicates like this without meaning to, and Google has a process for dealing with them called canonicalization. Understanding it helps you make sure the URL you want is the one that appears in results.

Where duplicate URLs come from

Duplicates are usually a by-product of how sites and servers work. Common sources:

• Protocol and host variants: http:// and https://, www. and the bare domain, all serving the same page.
• Tracking and session parameters: ?utm_source=, ?ref=, ?sessionid= added to links.
• Sorting and filter parameters: ?sort=price showing the same products in a different order.
• Trailing slashes and case: /shoes and /shoes/, or /Shoes and /shoes, both returning 200.
• Print versions, mobile URLs and syndicated copies on other sites.

None of these is a penalty risk on its own. The risk is practical: signals such as links get split across several URLs, crawling is wasted on copies, and Google may pick a version you would not have chosen.

What a canonical URL is

When Google finds a set of pages with the same or very similar content, it groups them and chooses one URL to represent the group. That is the canonical URL. Google usually shows only the canonical in results and consolidates signals onto it.

Google's choice is based on signals, and you control most of them.

The signals you can send

Redirects are the strongest signal. A permanent redirect (301 or 308) from a duplicate to your preferred URL tells Google and visitors that the duplicate should no longer be used. Use redirects when the duplicate does not need to exist for users, for example to send http:// to https:// and www. to the bare domain, or the other way round.

rel="canonical" is a link element in the page head that names your preferred URL. It suits duplicates that must stay accessible, such as tracking-parameter URLs or sort orders.

<link rel="canonical" href="https://example.com/shoes/">

Use an absolute URL, including the protocol and host. Relative URLs are easy to get wrong, and a malformed canonical tag is ignored. A page may point its canonical at itself; a self-referencing canonical is a common, harmless way of stating the preferred form of the URL.

Consistent internal links and sitemaps reinforce the choice. If your canonical tags say /shoes/ but every internal link and your sitemap use /shoes?ref=nav, you are sending conflicting signals, and Google may follow the majority.

Other things being equal, Google prefers HTTPS URLs over HTTP as canonicals.

Checking what Google chose

The URL Inspection tool in Search Console shows both the user-declared canonical and the Google-selected canonical for a URL. The Page indexing report lists pages under statuses such as "Duplicate, Google chose different canonical than user", which means Google weighed your canonical and picked another URL because other signals pointed there.

When that happens, compare the two URLs. Look at which one your internal links, redirects and sitemap favour, and whether the content really is the same. Fix the conflicting signals rather than repeatedly requesting indexing.

The myth of the duplicate content penalty

Many site owners fear a "duplicate content penalty". For ordinary duplication, such as parameter URLs or product descriptions shared with a reseller, there is no penalty. Google simply picks one version to show. Duplication becomes a problem under Google's spam policies only when it is deceptive or manipulative, for example scraping other sites' content at scale to rank with it.