Duplicate content: causes and fixes for business websites
Duplicate content rarely causes a penalty, but it splits signals and wastes crawling. Where it comes from on business sites and how to fix each cause.
By Sohail Hussain21 November 20265 min read
Duplicate content is the same or very similar content at more than one URL. On a business website it is almost always accidental: www and non-www versions, tracking parameters, tag archives, filtered category pages, a staging copy left open. Google does not penalise this. It picks one URL to show and folds the rest in. The cost is that Google may pick the wrong URL, links and signals are split across copies, and crawling is wasted. The fix is to give every page one URL: redirect the copies that should not exist, canonicalise the ones that must, and rewrite or merge pages that are too similar.
Duplicate URLs are one of the checks in our technical SEO checklist, and cleaning them up is a standard part of the indexing fixes in our technical SEO and performance work.
Is duplicate content a penalty?
No, with exceptions. Google's canonicalization documentation says some duplicate content on a site is normal and "not a violation of Google's spam policies". When Google finds a set of duplicates, it clusters them, chooses the one it considers most complete and useful as the canonical, and shows that one in results.
The exceptions are in Google's spam policies, and they are about intent:
- Scraping: republishing other sites' content without adding value.
- Doorway abuse: many pages or domains targeted at different cities or similar queries that funnel visitors to the same place.
- Scaled content abuse: generating large numbers of pages, by any method, mainly to manipulate rankings.
If none of those describe your site, duplication is a housekeeping problem, not a penalty risk.
Where does duplicate content come from on business sites?
| Cause | Example URLs | Fix |
|---|---|---|
| Protocol and host variants | http://example.com/about and https://www.example.com/about | 301 every variant to one host in one hop |
| Trailing slash or letter case | /about and /about/, /About | Pick one form; 301 the other |
| Tracking parameters | /pricing?utm_source=newsletter | Self-referencing canonical on every page |
| Sort and filter parameters | /products?sort=price&colour=blue | Canonical to the unfiltered page, or block in robots.txt on large catalogues |
| Tag, category and date archives | /tag/seo/ showing the same posts as /category/seo/ | Keep one taxonomy; noindex thin archives |
| Product in several categories | /shoes/runner and /sale/runner | One product URL; categories link to it |
| Staging or old site still live | staging.example.com, the old domain | Password-protect staging; 301 the old domain |
| Print or AMP versions | /article?print=1 | Canonical to the main version |
| Near-identical service or location pages | /plumber-pune, /plumber-mumbai with the same text | Rewrite with real local detail, or merge |
| Same content on two of your domains | example.com and example.in | Pick one, or use hreflang for genuine regional versions |
Most sites have three or four of these at once. Google's own list of typical causes, in the canonicalization overview, covers region variants, device variants, protocol variants, site functions such as sorting and filtering, and accidental variants such as a crawlable demo site.
How do you find duplicate content on your site?
Search Console's Page indexing report. Three reasons point straight at duplication: "Duplicate without user-selected canonical", "Duplicate, Google chose different canonical than user" and "Alternate page with proper canonical tag". The third is fine. The first two are worth a look.
URL Inspection. For any URL, it shows the canonical you declared and the one Google chose. When they differ, you have a cluster Google sees differently from you.
A site crawl. A crawler lists every URL it can reach from your links, including parameter and archive URLs you did not know existed, and flags pages with identical titles or near-identical text.
A quoted search. Copy a distinctive sentence from a page and search for it in quotes. Other URLs on your site, or other sites, that carry the same text appear in the results.
Which fix should you use for each cause?
Choose by asking whether the duplicate URL needs to exist.
- It should not exist (old domain,
http, wrong host, retired page): use a 301 redirect to the main version. It is the strongest signal and sends visitors to the right place too. - It must exist but should not rank (tracking parameters, sort orders, print versions, a product reached through two categories): use a canonical tag pointing to the main version. Canonical tags explained covers how to set them and why Google sometimes ignores them.
- It is useful to visitors but thin for search (tag archives, internal search results, paginated filters): use
noindex, or block crawling in robots.txt on very large sites where crawl volume is the problem. Do not use both on the same URL. - Two pages cover the same topic (two near-identical service pages written years apart): merge them into one stronger page and 301 the weaker URL. Our canonical tag generator helps for the ones you keep side by side.
- Regional versions in the same language (an English page for India and one for the UAE): keep both, add
hreflangand a self-canonical on each. Google recommends using both together for regional variants.
Search Console used to have a URL Parameters tool for telling Google which parameters to ignore. It was retired in 2022, so canonicals, consistent internal links and robots.txt are now the only levers.
After a fix, allow time. Google's canonicalization troubleshooting guide says pages can stay clustered for up to two weeks after you make them clearly different.
What about location pages and republished articles?
These are the two cases where duplication shades into a policy question.
Location pages. A service business that covers several cities often wants a page per city. That is legitimate when each page has something specific: the branch address and phone, the team, photos and examples of work in that area, local reviews, directions, prices that differ. It becomes doorway abuse when 30 pages share one template with the city name swapped. If you cannot write something real for a city, list it on one "areas we serve" page instead. Local SEO for service businesses in India covers how to structure these pages.
Republished articles. Publishing your post on LinkedIn, Medium or an industry site creates a copy on a stronger domain that can outrank you. Google's troubleshooting guide says the canonical tag is not recommended for syndication, because the pages are often too different, and that the most effective solution is for partners to block indexing of the copy. Where that is not possible, publish on your site first, let it be indexed, and make sure the copy links back to the original.
Once the copies are cleaned up, keep new ones from forming by linking only to the main version of every page, in menus, content and templates. Site architecture and internal linking covers how to set that up.
Was this guide useful?