Blog
Duplicate content is not a penalty, it is a choice Google makes for you.
The duplicate content penalty is one of the most durable myths in SEO, and it is worth killing because the fear it creates leads people to do genuinely harmful things: noindexing pages that should rank, panicking over trailing slashes, deleting content that was fine. There is no penalty. There is only Google, faced with two pages that say the same thing, choosing which one to show and quietly setting the other aside.
There is no penalty, and saying so matters
Google has stated plainly, more than once, that there is no penalty for duplicate content in the ordinary sense. Having the same text on two URLs does not get your site demoted or flagged. The only time duplication crosses into penalty territory is deliberate, large-scale scraping and spinning designed to manipulate, which is a different offence entirely and not what anyone worrying about a trailing slash is doing.
This matters because the myth is not harmless. Convinced they are being punished, people slap noindex on half their pages, canonicalise things that should not be canonicalised, and treat normal, benign duplication as an emergency. The fear causes more damage than the duplication ever would.
What actually happens
When Google finds two pages it judges equivalent, it clusters them and picks one, the canonical, to index and show. The others are not penalised, they are just set aside, and their signals are usually folded into the one it kept. This is Google being helpful: it is deduplicating the results so a searcher does not see the same page three times. Left to itself it is right most of the time.
The problem is the times it is wrong. If you do not tell it which URL you prefer, it guesses from signals like internal links and URL tidiness, and it can land on a tracking-parameter version, an http variant, or a print page instead of the one you would promote. That is the situation behind the Search Console status duplicate without user-selected canonical: not a penalty, but Google making a choice you should have made, and occasionally making it badly.
The real cost, and it is not a penalty
Two things genuinely cost you. First, Google may index the wrong URL, so the version that ranks is not the one you optimised. Second, on a large site, crawling thousands of near-identical parameter and filter URLs spends crawl budget that should have gone on pages that matter. Neither is a punishment. Both are inefficiency you remove by declaring the canonical yourself rather than leaving the machine to guess.
How to make the choice for it
The fix is to stop hoping and start declaring. Add a self-referencing canonical to every page so each one nominates itself, redirect true variants like http to https and non-slash to slash rather than only canonicalising them, and point your internal links at the canonical version consistently so the strongest signal agrees with the tag. For genuinely separate language versions, use hreflang rather than canonical, because they are not duplicates and should each rank in their own market. This is routine technical work, done once and maintained, not a rescue operation.
Where the duplication comes from thin, near-identical pages you created yourself, the answer is often not a canonical at all but consolidation: merge them into one stronger page, which is an on-page decision about what deserves to exist rather than a technical patch over pages that maybe should not.
The general lesson
A myth that makes people afraid is more dangerous than the thing they are afraid of, because fear drives overcorrection and overcorrection breaks working sites. Duplicate content is not a penalty. It is Google making a decision on your behalf, usually well, occasionally badly, and the entire job is to make that decision yourself so the machine never has to guess.
Keep reading
More from the blog
You are not being penalised, you are being outranked
A real penalty is rare and shows in Search Console. Most drops are something else.
Read itMeta descriptions do not affect rankings
Not a ranking factor, and Google rewrites most of them anyway.
Read itYou do not need a blog post for every keyword
A page per keyword splits your own authority. One strong page beats five thin ones.
Read itKeep reading
Related reading
Book me
Worried about duplicate content?
Send the domain. You get a straight read on whether Google is indexing the URLs you want, where the crawl budget is going, and whether there is anything real to fix, minus the penalty myth.