Duplicate content: what it is, what it costs, what it does not

The penalty myth

There is no duplicate-content penalty. When Google finds identical content at several URLs, it chooses one canonical representative and filters the rest out of results. The cost is dilution, not damage: links and signals split across the duplicates instead of concentrating on the version you wanted, and the chosen URL may not be the one your navigation, your sitemap or your ads point at.

Where store duplicates come from

• Template pages with thin differences: two products whose only difference is a size variant published as separate products; empty collections that render the same shell.
• Descriptions copied from suppliers: the same paragraph as hundreds of other stores: duplicate across the web, not just your site, which weakens product matching.
• Sort, filter and tag parameters multiplying one collection into many indexable URLs.
• Legacy pages: an old landing page and a new one both live after a redesign, answering the same query.

Which duplicates matter

Separate repeated metadata from repeated content. Two pages can share a title while describing different products; two URLs can have different titles yet serve the same article. Boilerplate shared across a template is normal. Review the useful main content and page purpose before choosing a structural fix.

The duplicate metadata finder compares exact titles and descriptions on sampled pages, ignoring letter case. It groups matching URLs; it does not crawl other websites for plagiarism or measure full-text similarity. A clear sample is not evidence that every page on your site is unique.

For a hypothetical store, imagine two distinct mugs imported with the title “Ceramic mug.” Keep both products and give each an accurate identifying title. By contrast, a campaign URL displaying the same product as its clean URL needs a canonicalization review. Those two observations lead to different actions.

Fix: consolidate or differentiate

1. Duplicate titles and descriptions: rewrite them per page. Use accurate distinctions from the page, such as model or purpose. A title rewrite can remove a metadata collision without changing the underlying content.
2. Genuinely duplicated pages: merge the content into one, 301 the rest, and point the sitemap at the survivor.
3. Duplicate URLs that must remain accessible: review a canonical pointing to the preferred equivalent. Do not automatically canonicalize distinct product variants to a parent.
4. Verify after fixing: inspect the changed metadata and declared canonical, then compare the same sampled URLs again. Search index changes need later evidence.

Google can use a meta description or page content to generate a snippet, so a unique description does not guarantee the text shown in results. See Google’s snippet guidance. For duplicate or very similar pages, review its canonicalization guidance and the worked canonical guide.

Run the free duplicate-content check to see which titles and descriptions collide on your site today. Related: fixing keyword cannibalization · canonical check

Questions

Keep reading