Keyword cannibalisation happens when two or more pages on the same website aim at the same search and the same need, so they compete with each other in Google’s results. Instead of one strong page, the site offers several partial ones, and Google has to guess which to show.
How keyword cannibalisation works
Google generally shows only one or two results from a single site for a given query. When several of your pages look like plausible answers, it may rotate between them, rank a weaker one, or rank none of them highly, because links, engagement and relevance are split across pages rather than concentrated on one.
The cause is usually growth without a plan. A blog post from 2021 on “how much does a loft conversion cost”, a 2023 post on “loft conversion prices in London” and a service page with a cost section all chase the same question. Tag and category archives listing the same posts, product variants on separate URLs and old campaign landing pages that were never removed create the same overlap.
Not every shared keyword is cannibalisation. A service page for “boiler installation” and a guide on “how long does a boiler installation take” share words but serve different search intents, and both can rank. The test is whether the pages answer the same need for the same person.
Why it matters
For a small UK business with limited authority, splitting effort across duplicate pages is expensive. Every link, every update and every internal link goes to whichever page someone happened to pick, so none gains enough to compete with a rival who has one clear, thorough page.
It also confuses visitors and muddles reporting. A searcher may land on an outdated 2021 article with old prices instead of the current service page, and in Search Console the clicks for one topic are scattered across URLs so trends are hard to read.
Common mistakes
- Writing a new post instead of updating the old one. The most common cause of cannibalisation on blogs.
- Deleting a page without redirecting it. Any links and visibility it had are lost rather than passed to the page you keep.
- Calling any shared word cannibalisation. Pages with different intents are not competing, and merging them can hurt both.
- Using a canonical tag as the default fix. A canonical is a hint, and it is meant for near-duplicates. For two distinct articles, merging is usually cleaner.
- Linking internally to both pages with the same anchor. This tells Google you cannot decide either.
How to act on it
Open the Performance report in Google Search Console, filter by a query you care about, and switch to the Pages tab. If several URLs share the impressions and the ranking page changes from week to week, look closer. A site: search with the topic can reveal pages you had forgotten.
For each overlap, decide which page best fits the intent. Then choose a fix:
- Merge: combine the useful content into the stronger page and point the others to it with a 301 redirect. This kind of content consolidation is the most common fix.
- Differentiate: rewrite one page so it targets a genuinely different need, such as turning a second cost post into a guide on financing.
- Canonicalise: for near-identical pages that must both exist, such as product variants, use a canonical tag to name the preferred one.
Then update internal links to point at the page you kept. Prevention is simpler than cure: check every planned page against existing ones before writing it. Untangling overlapping content and planning one page per topic is part of the content SEO and strategy work I do.