Duplicate Content: Why Your Site Creates It on Its Own — and How to Stop It
Most duplicate content was never written by anyone: your own website manufactures it with www and non-www versions, URL parameters and product copy pasted from the manufacturer. Google doesn't fine you for it — it does something worse: it splits your authority across copies.
November 1, 20236 min readIn this article
- What it is — and the penalty myth
- The duplication your site manufactures by itself
- The duplication you do write yourself
- The canonical tag: telling Google which one counts
- Languages: a translation isn't duplication — if you declare it
- How to find the duplication you already have
- And when someone else is the one duplicating
- Frequently asked questions
When people hear "duplicate content" they picture a competitor stealing their copy. Reality is less cinematic: most duplication is generated by your own website without anyone writing a word. The www and non-www versions, the URL with campaign parameters, the product page pasted straight from the manufacturer's catalogue — all of it is copies competing against each other.
This guide explains what duplicate content actually is, why the famous "penalty" doesn't exist (what really happens is subtler and still costs you traffic), how to find the duplication you already have, and how to fix it with three tools: redirects, canonicals and hreflang.
What it is — and the penalty myth
Duplicate content is the same text (or nearly the same) reachable at more than one URL. Google doesn't fine you for that: it picks one version, shows it, and ignores the rest. The problem is that it sometimes picks the one you didn't want, and that links pointing at the copies split their strength instead of stacking it on a single page. It isn't punishment; it's dilution. The practical outcome is the same: you rank lower than you could.
The real exception is large-scale duplication built to manipulate — entire sites scraped, cloned domains. That can trigger a manual action. A small business with repeated product copy or two versions of its homepage is nowhere near that territory, so work calmly: this gets fixed with housekeeping, not panic.
The duplication your site manufactures by itself
Before you audit your copy, audit your URLs. These are the usual technical sources — each one turns a single page into two, four or eight copies:
| Source | Example | Fix |
|---|---|---|
| www vs non-www, http vs https | http://yoursite.com and https://www.yoursite.com | 301 redirect to a single canonical version |
| Trailing slashes and index pages | /services and /services/ | 301 redirect or canonical to one form |
| URL parameters | /shoes?colour=black&utm_source=meta | Canonical to the clean URL |
| Store filters and sort orders | /shirts?sort=price creates a "page" per combination | Canonical to the base category; keep combinations out of the index |
| Print versions or leftover AMP pages | /product/print | Remove them, or canonical to the main version |
The duplication you do write yourself
The second family is editorial, and it's the most common one in online stores: the manufacturer's description pasted verbatim. If you and forty other shops use the same catalogue paragraph, Google has forty-one identical copies and no reason to show yours. Rewriting the pages of your twenty best-selling products — with real measurements, usage context and your own photos — beats any technical trick.
- Product copy from the manufacturer's catalogue: rewrite at least your best sellers.
- Cloned city pages ("web design in Quito", "web design in Cuenca") where only the place name changes: either they carry real local content, or they're duplication under another name.
- The same promotion published as a page, a blog post and a press note: pick one destination and link to it.
- Legal and shipping text repeated on every product page: better a short block linking to the full page.
The canonical tag: telling Google which one counts
The `rel=canonical` tag is one line in the `<head>` declaring the official URL of a piece of content. It doesn't stop variants from existing — that's what redirects are for — but it tells Google where to consolidate signals when variants are unavoidable, like store filters or campaign parameters.
The rule for choosing your tool is simple: if the duplicate URL shouldn't exist for anyone, use a 301 redirect; if it should exist for the user but not compete on Google, use a canonical. And one frequent mistake worth checking: pages whose canonical points at a different page through a template slip — quietly telling Google "don't show me" without meaning to.
Languages: a translation isn't duplication — if you declare it
A website in Spanish, English and German isn't duplicate content: it's the same message for different audiences. But Google needs you to declare it with hreflang tags, which connect each version to its equivalents. Without them, two versions in the same language aimed at different countries can end up competing with each other. When several languages are worth having, and how to maintain them, is covered in multilingual websites: when they make sense.
How to find the duplication you already have
- Search Google for `site:yoursite.com` and look for parameter URLs, repeated versions or pages you don't recognise.
- Take a distinctive sentence from an important page, search it in quotes, and see how many of your URLs (and other people's) return it.
- In Search Console, check Indexing → Pages: the "Duplicate" and "Google chose different canonical" statuses hand you the exact list, no guessing.
- In a store, check how many product pages share the manufacturer's description: one quoted search per product is enough.
And when someone else is the one duplicating
When another site copies your copy, Google almost always identifies the original correctly — especially if your page has been indexed for a while. If the copycat outranks you, you have two routes: ask them directly to take it down (works more often than you'd think) or file a copyright complaint with Google. What's not worth doing is rewriting your own content to "differentiate yourself from your copy": strengthen the original with internal links and authorship signals, as part of a complete SEO strategy.
Frequently asked questions
Does Google penalise duplicate content?
Not in the sense of a fine: Google filters the copies and shows a single version. The real "penalty" is indirect — authority split across variants, and possibly the wrong version being shown. Manual actions are reserved for mass duplication built to manipulate, not for a small business with untidy URLs.
How much duplicate content is acceptable?
There's no magic percentage. Unavoidable repeated blocks — menu, footer, legal notices — are normal and Google understands them. The problem starts when the main content of two URLs is essentially the same. The useful question isn't "how much repeats?" but "does this page offer something the other doesn't?".
Should I use a 301 redirect or a canonical?
A 301 redirect when the duplicate URL shouldn't exist at all: old versions, http, non-www. A canonical when the variant must keep working for the user but shouldn't compete on Google: store filters, campaign parameters. In doubt, the redirect is the stronger tool — a canonical is a suggestion Google may ignore.
Does translating my website into English count as duplicate content?
No. A translation is distinct content to Google. The risk appears with two versions in the same language aimed at different countries (Spain and Ecuador, say): that's where you need hreflang so each one is shown to its audience instead of competing with the other.