Duplicate content: what it is, whether it hurts SEO, and how to fix it

Duplicate content is the same or nearly the same content reachable at more than one URL, either on your own site or across different sites. It is very common, and Google says some duplicate content on a site is normal and not a violation of its spam policies, so there is no penalty for it. What it does cost is control: Google groups the copies, picks one URL to show, and may pick the wrong one, while links and crawling are split across the copies. This guide explains what counts as duplicate content, where it comes from, how to find it, and how to consolidate it with redirects and canonical tags. The tag itself is covered in canonical tags explained, the main guide for this topic.

Quick answer

• Duplicate content is normal. Parameters, www and non-www, http and https, print versions and products in several categories all create it.
• There is no duplicate content penalty for ordinary duplication. Google acts only when copying is deceptive or manipulative, such as scraped or mass-produced pages.
• The real costs: Google may show a URL you did not want, links and shares are split across copies, and crawling is spent on duplicates.
• Fix each source at the root: 301 redirects for host and protocol variants, a canonical tag for parameters and variants people still use, merging or rewriting for near-duplicate pages.
• Do not block duplicates in robots.txt. Google cannot read the canonical of a page it may not crawl.
• Check what Google chose with URL Inspection in Search Console, and your own signals with the canonical checker.

What counts as duplicate content?

Google uses "duplicate" for pages that are identical or so similar that they answer the same need. That covers more than copied text:

• Exact duplicates: the same page at two addresses, such as https://example.com/shoes/ and https://www.example.com/shoes/.
• Near-duplicates: pages that differ only in a small part, such as a product list sorted by price, the same product in red and blue with the same description, or city pages that swap only the town name.
• Cross-domain duplicates: your article republished on a partner site, or copied by a scraper, or a manufacturer's product description used by hundreds of shops.

Two things are usually not duplicate content: the same content translated into another language, and boilerplate such as the header, footer and navigation that repeats on every page. Google recognises template text and looks at the main content.

Is there a duplicate content penalty?

No, not for normal duplication. Google's documentation on canonicalization says "some duplicate content on a site is normal and it's not a violation of Google's spam policies". Google has said the same since its 2008 Webmaster Central post "Demystifying the 'duplicate content penalty'".

What Google does act on is deception. Its spam policies cover scraped content (copying others' pages without adding value) and doorway pages (many near-identical pages made to rank for variations of a search). Those can lead to lower rankings or a manual action. A shop with sorted category URLs or a blog with a print version is nowhere near that line.

What duplicate content really costs

Duplication does not lower your rankings as a punishment, but it has side effects that can look like one:

• Google picks the URL. It groups duplicates into a cluster and indexes one as the canonical. If you have not said which one, it may choose the parameter version, the http version or the copy on another site.
• Signals are split. Links and shares to three versions of a page are three weaker signals until Google consolidates them, and it only consolidates what it recognises as the same.
• Crawling is spent on copies. Google crawls non-canonical duplicates less often, but on a shop with thousands of filter URLs, those requests still come at the cost of new pages.
• Near-duplicates compete. Two similar pages targeting the same search can take turns ranking, and neither ranks as well as one strong page would.
• Reports get noisy. Search Console fills with "Duplicate" statuses, which hides real problems.

Where duplicate content comes from

For parameters and filters in detail, see faceted navigation and parameter URLs. For the host and slash variants, see trailing slashes, www and https: pick one.

How to find duplicate content on your site

Search Console's Pages report is the best source because it shows what Google itself grouped. Three reasons are about duplicates:

• "Duplicate without user-selected canonical": Google found a duplicate with no canonical tag and picked another URL. What that status means and how to fix it goes through it case by case.
• "Duplicate, Google chose different canonical than user": your canonical says one URL, Google picked another.
• "Alternate page with proper canonical tag": a duplicate correctly points at its original; usually no action.

Open the examples in each and look for a pattern: a parameter, a path, a template.

URL Inspection on a single URL shows the user-declared canonical and the Google-selected canonical side by side.

A quoted search. Copy a distinctive sentence from a page and search for it in quotes. Other URLs on your site with the same text are internal duplicates; other sites are copies or syndication.

A crawl. A desktop crawler lists pages with identical titles, descriptions or body text across the site. Duplicate titles are often the first visible symptom; see duplicate titles and descriptions across a site.

How getReport checks it

The canonical checker runs a report on one URL and shows the chain from the address you entered to the URL Google should index: the redirects, every canonical tag in the HTML, whether the canonical answers 200, and whether the http and www variants redirect to it or serve their own copy. "Serves its own copy" is a duplicate you can fix at the server today. Paste a parameter version of a URL to see where its canonical points.

It checks one page and its variants per run, so it confirms a pattern rather than finding every duplicate on a large site.

How to fix duplicate content

Pick the fix by whether visitors need the duplicate:

1. Nobody needs it: redirect. http, www, slash and case variants, old URLs, merged pages. A 301 is the strongest signal and moves visitors too. See 301 redirects.
2. Visitors use it, but it should not be indexed: canonical. Tracking and sort parameters, print versions, products under a second category path. Add <link rel="canonical" href="…"> pointing at the original, and link internally to the original only.
3. The pages are similar but meant to be different: make them different. City pages, variants, near-identical articles. Add what is specific to each, or merge them into one stronger page and redirect the rest.
4. The page has no search value at all: noindex. Internal search results, thin archives. Google's documentation says not to use noindex to choose a canonical between duplicates, because it drops the page and its signals rather than merging them.
5. Keep the signals consistent. Internal links, the XML sitemap, hreflang and og:url should all use the canonical URL. When they disagree with your tag, Google may choose a different canonical.

Near-duplicates: location pages, variants and thin archives

Near-duplicates are harder than parameters because no tag fixes them. The question is whether each page gives a searcher something the others do not.

• Location pages: "Plumber in Leeds" and "Plumber in York" with the same 600 words and a swapped town name look like doorway pages. Keep a page per place only where you have something specific: the address, the team, local jobs, prices, reviews. Otherwise one page listing the areas you serve is better.
• Product variants: the same shoe in six colours with six identical descriptions. Either one URL with a selector, or distinct content per variant.
• Archives: tag, date and author archives on a blog often list the same posts. Keep the useful ones indexable with an intro; noindex the rest.

Short pages are not duplicates by definition; thin content: when short pages are fine explains when length matters.

Duplicate content on other sites

When your content appears on another domain, Google picks one version to show, usually the one it considers the original or most useful.

• Syndication you agreed to: Google now recommends that the partner uses noindex on its copy, because a cross-domain canonical to a page with different surroundings is often ignored. Canonicals across domains has the details.
• Scrapers: usually nothing to do; Google is good at picking the original. If a copy outranks you, a copyright removal request to Google is the formal route.
• Your own second domain: redirect it with 301s rather than serving a copy.

Common mistakes

• Treating every "Duplicate" status as an error. Alternates with a correct canonical are the system working.
• Canonicalising near-duplicates that should be different pages. The canonical says "these are the same"; if they are not, Google may ignore it.
• Using noindex and a canonical together on the same duplicate. One says drop, the other says merge.
• Rewriting a sentence or two on location pages to make them "unique". Google looks at whether the page is useful, not whether a string differs.
• Internal links to parameter URLs, such as tracked links in your own menu. They tell Google the duplicate matters.

Questions people ask

What is duplicate content in SEO?

Duplicate content is the same or very similar content reachable at more than one URL, on one site or across sites. Common causes are URL parameters, http and https or www and non-www versions, print pages, products in several categories and copied text. Google groups the copies and indexes one of them as the canonical, which is why you should tell it which URL you prefer.

Does duplicate content hurt SEO?

Not as a penalty, but it can still cost you. Google does not punish normal duplication, yet it may show a URL you did not choose, split links and shares between copies, and spend crawling on duplicates instead of new pages. Near-identical pages can also compete for the same search. Redirects, canonical tags and consolidating similar pages remove those costs.

Is there a duplicate content penalty from Google?

No, not for ordinary duplication. Google's canonicalization documentation says some duplicate content on a site is normal and not a violation of its spam policies. Google only acts when copying is deceptive, such as scraped content or doorway pages made to rank for many variations of a search, which can lead to lower rankings or a manual action.

How much duplicate content is acceptable?

There is no percentage limit. Repeated navigation, footers and short legal text on every page are expected and ignored. What matters is the main content: if two pages would give a searcher the same answer, treat them as duplicates and consolidate them. If each page has its own purpose and substance, some shared text between them is fine.

What should I do if another site copies my content?

Usually nothing, because Google is good at identifying the original and showing it. Make sure your page was published first, is indexed and has a self-referencing canonical. If the copy outranks you or harms your business, contact the site owner, then file a copyright removal request with Google. For partners you allow to republish, ask them to add noindex to their copy.