Duplicate Content: Fix the Confusion, Skip the Panic
There is no penalty for this. There is confusion, split strength, and Google choosing the wrong version of your page.
Why duplication costs you without a penalty
Google picks one version and ignores the rest. If it picks the wrong one, you lose without ever being punished.
-
One page gets published
You write it once and it goes live at a single address, as far as you know.
-
It becomes several addresses
With www and without, with a trailing slash and without, plus filters, archives and print versions. All the same page.
-
One version gets declared
The real address is chosen and stated consistently in canonicals, internal links and the sitemap, and the rest stop being created.
There is no duplicate content penalty. That is worth saying first, because most of the panic about this is sold rather than earned.
What there is, is confusion. When the same content sits on several addresses, Google picks one and ignores the rest. If it picks the wrong one, or splits your strength across several, you lose without ever being punished.
The Honest Version: Most Duplication Is Harmless
What does not matter
Your boilerplate. Shared headers, footers, disclaimers and product specifications repeated across pages are completely normal. Every site has them and no site is penalised for them.
Quoting a paragraph from somewhere else with attribution is fine too.
What genuinely costs you
The same page reachable at several addresses, so your links and strength split between them. Printer versions, tag and category pages that reproduce full articles, filter and sort URLs, and the same page with and without a trailing slash.
And the one nobody checks
A page duplicating itself. Broken markup can make a template output the same content twice on one URL. It looks fine to a reader scrolling, and it doubles the weight of the page.
Where It Comes From
Almost always a technical setting rather than a writer.
Several addresses, one page
With and without www, http and https, with and without a trailing slash. Four versions of everything.
Filters and sorting
Every combination generating a new address with the same products on it.
Tag and category pages
Reproducing whole articles instead of excerpts, so the archive competes with the article.
Printer and AMP versions
A second copy of every page, often forgotten and left indexable.
Content repeated across your own pages
Location pages built by swapping one word. Google has discounted this for years.
A page duplicating itself
Broken template markup outputting the content twice. Invisible unless you look at the source.
How It Gets Fixed
1. Find every address that serves the same thing
By requesting them, not by reading settings. Most duplication is discovered by trying the variations rather than by auditing a list.
2. Decide which version is the real one
Then say so consistently through canonical tags, internal links and the sitemap. All three have to agree, and often they do not.
3. Stop the rest being created
Parameters handled, filters kept out of the index, archives switched to excerpts. Fixing the symptoms without stopping the source means doing it again in six months.
4. Check the page against itself
Confirming the page outputs its content once. This takes a minute and almost nobody does it.
A Real Example
On this site, one of the highest-traffic pages was serving its entire article twice on a single URL.
The database held one copy. The page rendered two, because the markup contained broken block tags that confused the template into outputting the content again.
Nobody had noticed. It looked normal to read, and no audit tool flagged it, because tools compare pages against each other rather than a page against itself.
Removing the broken markup cut the page from 150KB to 120KB and left one clean copy.
When To Ignore It
It is boilerplate. Shared headers, footers and legal text are not a problem and never have been.
A tool gave you a percentage. Similarity scores measure words, not intent. Two pages can share vocabulary and serve completely different searches.
Google is already picking correctly. If the right version ranks and the others are ignored, the system is working. Leave it.
What You Can Check Yourself First
Most duplication is found by trying addresses rather than running a tool.
1. Try the four versions of your homepage
Load your site with and without www, and with and without https. All four should end up at the same single address. If more than one loads without redirecting, you have four versions of every page on your site.
2. Add a trailing slash and see what happens
Load one of your pages with a slash on the end and then without. If both load and neither redirects, that is another duplicate for every URL you own.
3. Check what Google chose
Paste a page into the URL Inspection tool in Search Console. It reports the canonical Google selected. If that is not the address you expected, you have found a real problem and it is costing you.
4. Search a distinctive sentence from your own page
Take an unusual sentence from a page, put it in quotes, and search site:yourdomain.com with it. If more than one of your URLs comes back, that content exists in more than one place on your own site.
Price and Turnaround
Scope |
Price |
Turnaround |
|---|---|---|
Standard site |
$249 |
3 – 5 days |
Ecommerce or filtered catalogue |
$549 |
1 – 2 weeks |
Large or multi-domain |
From $899 |
2 – 3 weeks |
Includes finding every duplicate address, choosing and declaring the correct version, and stopping the source so it does not rebuild.
Common Questions
Do I need to worry about scrapers copying my content?
Rarely. Google is generally good at identifying the original, particularly if yours was indexed first. Chasing scrapers is usually wasted effort.
Should every page have a canonical tag?
Yes, and on most pages it should point at itself. That removes ambiguity rather than creating it.
What about pagination?
Page two of a listing is not a duplicate of page one and should not be canonicalised to it. That was old advice and it loses the pages beyond the first.
Does an SSL certificate cause this?
Not the certificate, but moving to https without redirecting the http version does. That is where the four-versions problem usually starts.
Is there a duplicate content penalty?
No. There is a filter that hides duplicates and a canonical choice that can go against you. Those cost you traffic without being a penalty.
Does copying my own product descriptions hurt?
Across your own site, rarely. The bigger issue is if every retailer uses the same manufacturer text, which leaves nothing to distinguish you.
What about location pages?
If they differ only by the place name, they are duplicates and get discounted. Genuinely different pages for genuinely different markets are fine.
Can I just add canonical tags everywhere?
Only if they point at the right page. Wrong canonicals are worse than none, and I find them regularly.
How do I know Google picked the wrong version?
Search Console tells you which URL it chose. If it is not the one you wanted, that is the signal.
Not sure which version Google kept?
Send me a page and I will tell you free how many addresses serve it and which one Google chose.
Or email info@shazzseo.com