Decide which version should exist, then redirect or canonicalise the rest. Merge pages that compete for the same search.
For sites with similar pages, parameter URLs or competing posts. You will leave with one URL per idea and clean signals.
Duplicate pages are not a penalty in themselves. The cost is quieter: search engines have to guess which version to show, links and signals get split across copies, and your own pages end up competing with each other. Cleaning them up concentrates what you have onto one strong URL.
1Find the duplicates
Crawl the site with a tool such as Screaming Frog or Sitebulb and group pages by title, content similarity and URL pattern. Then check the structural sources that create copies without anyone writing a thing.
| Source | Example |
|---|---|
| Hostname and protocol | http://, https://, www and non-www all load the page |
| Trailing slashes | /pricing and /pricing/ both return 200 |
| Parameters | ?utm_source=, ?ref=, ?sort=price |
| Filters and sorting | /shoes?colour=red&size=9 |
| Printer or AMP versions | /page/print |
| Overlapping posts | Two articles for the same search |
2Choose the keeper
For each cluster, pick the version that should survive. Decide using links, traffic and quality, not which is oldest.
- Which has more links from other sites?
- Which gets more clicks and impressions?
- Which is more complete and accurate?
- Which has the cleaner URL?
3Pick the right fix for each situation
| Situation | Fix |
|---|---|
| The same page at two URLs, one is not needed | 301 redirect the extra one to the keeper |
| A variant that must exist, such as a filtered view | Canonical tag pointing to the main page |
| Two articles competing for one search | Merge into the stronger one and redirect |
| Tracking parameters | Canonical to the clean URL |
<link rel="canonical" href="https://example.com/shoes" />location = /old-guide {
return 301 https://example.com/guides/new-guide;
}4Merge overlapping pages properly
When two posts cover the same search, merging is better than rewording one. Take whatever the weaker page does well, such as an example or a section the other lacks, fold it into the keeper, then redirect the old URL.
Before
"How to reduce churn" and "Churn reduction tips" both target the same search, each with some links and thin traffic, and neither ranks well.
After
One complete guide with the best of both, the weaker URL 301-redirected to it, internal links updated to point at the survivor.
5Update the signals and recrawl
- Change internal links so they point at the final URL, not a redirect
- Remove redirected and non-canonical URLs from the sitemap
- Check each keeper has a self-referencing canonical
- Recrawl and confirm one URL per idea
- Watch Search Console for the old URLs dropping out and the keeper's impressions holding
Worked example
Cleaning up a retail blog (illustrative)
A made-up retail analytics tool finds its blog loads at both www and non-www, with tracking parameters creating hundreds of near-copies, and three posts about the same question. It redirects non-www to www in one hop, sets canonicals to clean URLs, merges the three posts into one and updates the sitemap. A month later Search Console shows fewer duplicate and alternate-page entries, and one guide ranks where three had split the traffic.
Say it in 30 seconds
The 30-second version
Find the clusters, choose the strongest page in each as the keeper, and redirect true duplicates to it with a 301. Canonicalise variants that must exist, update internal links and the sitemap, and recrawl to confirm.
Worksheet
Duplicate content cleanup sheet
The full guide and a worksheet to fill in, as a PDF you can print or share with your team. The worksheet covers:
- Crawl the site and group pages by title, content and URL pattern
- Test hostname and protocol variants: http, https, www and non-www
- List parameter and filter URLs that create copies
- For each cluster, name the keeper and the reason
- and 6 more
One email with the PDF. No list, no follow-ups.
