>

Duplicate Content and Canonicalisation

Duplicate content is rarely malicious and rarely penalised. It is usually a structural accident that splits ranking signals across multiple URLs and quietly costs visibility.

How Duplicate Content Actually Arises

Almost always through URL variations serving the same content:

  • HTTP and HTTPS versions both resolving
  • www and non-www both resolving
  • Trailing slash and non-trailing-slash variants
  • Tracking parameters creating infinite unique URLs
  • Faceted navigation generating combinations
  • Printer-friendly versions and AMP pages
  • The same page reachable at /page.html and /page/

Syndicated content and boilerplate product descriptions copied from manufacturers are the content-level versions of the same problem.

What Google Actually Does

There is no general duplicate content penalty. Google selects one version it considers canonical and consolidates signals onto it.

The cost is that Google may not choose the version you want, links split across variants dilute authority, and crawl budget is spent on duplicates rather than new content. On large sites the crawl waste is the material problem.

Canonical Tags

A canonical tag tells search engines which version is preferred. It is a strong hint, not a directive — Google may disregard it if other signals conflict.

Common mistakes: canonicals pointing to redirecting URLs, relative rather than absolute URLs, every page canonicalising to the homepage, conflicting canonical and hreflang, and paginated series all canonicalising to page one.

Self-referencing canonicals on every page are good practice. They remove ambiguity when parameters are appended.

Choosing the Right Fix

Match the mechanism to the situation:

  • 301 redirect when one version should not exist at all — HTTP to HTTPS, old URL to new
  • Canonical tag when both versions need to remain accessible but only one should be indexed
  • noindex when a page must exist for users but has no search value — internal search results, thin filter combinations
  • robots.txt to prevent crawling entirely; note this also prevents Google seeing a canonical or noindex on those URLs

When Google Ignores Your Canonical, and Why

A canonical tag is a hint, not a directive. Google selects a canonical itself and will override yours. Understanding the override conditions saves a great deal of wasted diagnosis.

The common reasons a declared canonical is disregarded: the target differs substantially in content, so the pages are not duplicates and the declaration is simply wrong; conflicting signals, where internal links, the sitemap and hreflang point somewhere else than the tag does; the canonical target is non-indexable, whether noindexed, blocked in robots.txt or returning an error; canonical chains, where A points to B which points to C; and a relative URL that resolves wrongly on some templates.

The diagnostic that answers it directly is the URL inspection tool, which reports the user-declared canonical and the Google-selected one side by side. When they differ, the problem is a signal conflict somewhere else on the site, and changing the tag again will not fix it.

Choosing the Right Instrument for the Problem

Canonical tags are the default answer to duplication and frequently the wrong one. Each instrument does something different and they are not interchangeable.

SituationInstrument
Two URLs, genuinely the same pageCanonical to the preferred one
A page permanently moved301 redirect, not a canonical
Faceted or filtered variants with no search valuenoindex, or block the parameter
Paginated sequencesSelf-referencing canonicals on each page
Same content for different regions or languageshreflang plus self-referencing canonicals
Syndicated to a third partyAsk for a canonical back to you, or accept the risk

Two errors are worth naming because they are frequent and expensive. Canonicalising paginated pages to page one removes deeper items from consideration entirely. And combining noindex with a canonical on the same URL sends contradictory instructions — one says do not index this, the other says consolidate its signals into another page — and the outcome is unpredictable. Choose one.

Sources

What each claim on this page rests on. Entries are typed so you can see which are primary.

  1. officialGoogle Search Central documentation — canonicalisation, duplicate handling, and how Google selects a canonical URL developers.google.com

Ask an AI about this page

Opens your assistant with this page as the source, and a question rather than a summary. It will ask what you are building before it answers.

ChatGPTClaudeGeminiPerplexityGrok

Nothing is sent from here. The link carries only this page’s title and address.