What is keyword cannibalization?
Short answer
Keyword cannibalization is when two or more pages on the same website target the same search, so they compete with each other. Google has to choose one, and may pick the weaker page or switch between them. It isn't a penalty. The fix is to merge the pages or give each its own search.
When two or more pages on the same website target the same search, they compete with each other and not with your competitors, and that is keyword cannibalization. Google then has to choose which one to show. It may choose the weaker page, switch between them, or rank both lower than one clear page would rank. The term comes from the SEO industry, not from Google. It describes a muddle a site creates for itself, and no penalty is involved.
What it looks like on a home-service site
Take a hypothetical HVAC company in El Cajon with three pages: a service page for AC repair, a blog post titled “AC repair in El Cajon: what to expect”, and an El Cajon city page whose text is mostly about AC repair. All three are aimed at the same search. Links from other sites and links within the site are spread across three addresses, and no single page makes the full case.
Sites rarely plan this. It builds up when pages are added on a schedule without checking what already exists.
When two pages for one search are fine
Overlap is a problem only when it costs you something. If two of your pages both appear on the first page for the same search, leave them alone.
Pages can also share words without competing. Water heater repair and water heater replacement use much of the same vocabulary, but the customer wants a different thing in each case, and Google generally treats them as different searches. What a page is for matters more than which words it contains.
How to check for it
Google Search Console can show it, one search at a time. The labels in the tool change now and then, but the steps are roughly these:
- Open the performance report for search results.
- Filter it to one search that matters to you.
- Switch to the list of pages. It shows every page of yours that appeared for that search.
- If there’s more than one, look at them over several months. Pages that take turns, with none of them holding a steady position, are the sign.
A quicker, rougher check is to list your page titles and main headings in one column and look for near-twins.
How to fix it
Whether both pages have a reason to exist decides what to do.
| Situation | Fix |
|---|---|
| Two pages say much the same thing | Move the best parts into the stronger page, then redirect the other address to it permanently. |
| The pages overlap, but each has a real purpose | Rewrite the title, heading and opening of each so it targets its own search. |
| The right page exists, but Google shows the wrong one | Point the links within your site at the right page, using words that describe it. |
| Near-identical pages must both stay | Add a canonical tag, a line of code that tells Google which version to treat as the main one. |
Deciding which search each page owns is part of keyword research. Keeping a map of one search per page, so the problem doesn’t come back, is part of site architecture. I fix one pair at a time and measure about two weeks later, so it’s clear whether the merge helped.
How it differs from duplicate content
Cannibalization is different pages chasing the same search. Duplicate content is the same text sitting at more than one address, and it calls for different fixes.