Insights — Digital Infrastructure — 6 min read
What Is Keyword Cannibalisation? When Your Own Pages Compete With Each Other
Keyword cannibalisation is when two or more of a company's own pages target the same search intent closely enough that they compete with each other instead of a competitor — weakening both rather than strengthening either.
In short
Keyword cannibalisation is when two or more pages on the same website target the same search intent closely enough that they compete with each other for ranking, rather than each owning a distinct question. The result is usually that neither page ranks as well as a single, clearer page would. It is a real and documented risk, but genuinely related pages that answer different questions — a service page and a background article, for example — do not cannibalise each other; they can support one another through internal links.
A company builds a service page. Then an article answering a closely related question. Then another article covering what looks like slightly different ground. A year later, nobody is sure which page is supposed to rank, and Search Console shows two or three URLs quietly taking turns appearing for the same query — none of them doing particularly well. That pattern has a name: keyword cannibalisation.
It is a genuine, documented problem, but it gets over-diagnosed. Not every pair of similar pages is cannibalising each other, and the fix is not always to delete something. This article sets out what cannibalisation actually is, when it is real, when it is not, and what to do about it.
What keyword cannibalisation actually is
- Keyword cannibalisation
- An industry term (not a phrase Google uses in its own documentation) for the situation where multiple pages on one site are relevant enough to the same search query that a search engine has to choose between them, splitting ranking signals that would otherwise concentrate on a single, stronger page.
Google's own documentation does not describe 'cannibalisation' as a named penalty. What it does document is the related, underlying problem of duplicate or near-duplicate content: when multiple URLs are similar enough, Google has to pick a canonical version to show, and signals such as links and relevance get divided across the versions rather than consolidated on one. Google's guidance on consolidating duplicate URLs addresses exactly this mechanism, even though it does not use the word cannibalisation.
Can two of my own pages really compete with each other?
Yes, and it happens more easily than most businesses expect. It is not only exact duplicates — two pages can be written completely differently and still cannibalise each other if they answer the same underlying question for the same searcher at the same stage of their decision. A page titled 'Commercial Website Audit' and an article titled 'What Should a Commercial Website Audit Include' can end up targeting the same query, even though one is a service page and the other is informational in tone.
When similar pages genuinely compete vs when they support each other
The distinguishing question is not how similar the topics sound, but whether a single searcher typing one query would be well served by either page interchangeably. If yes, they are competing. If the pages serve different intents — one explains a concept, the other sells a service, a third compares options — they are not competing; they are different stops on the same buyer's journey, and should link to each other rather than be merged.
| Signal | Genuinely competing (cannibalisation) | Legitimately complementary |
|---|---|---|
| Search intent | Same query, same intent, same stage of decision | Related topic, different intent or decision stage |
| Content overlap | Say the same thing in different words | Cover different depth, angle or next step |
| Search Console | Both URLs alternate appearing for the same query, both with weak position | Each URL has its own distinct query set |
| Buyer's use | Reader would be equally served by either page | Reader needs one, then naturally wants the other |
| Right action | Consolidate, differentiate clearly, or redirect | Keep both, link between them with clear anchor text |
How service pages and articles should interact
A service page and a related article are not naturally in competition, because they typically serve different intents: the service page states what Evans does, for whom, and how to buy it; the article answers a question a prospective buyer has on the way to that decision. The practical risk is not that they compete, but that they are built so similarly — same keyword in the same position, same generic claims, no distinct job — that a search engine cannot tell them apart. The fix is to keep the service page commercial (what it is, proof, how to start) and let the article do the explaining, then link clearly from the article to the service page and from the service page to one or two of the most relevant articles.
Should I delete old articles?
Sometimes, but deletion is a blunt instrument and should not be the default response to a suspected overlap. Before deleting, check whether the article still receives impressions or clicks in Search Console, whether it covers a genuinely distinct angle that simply needs sharper differentiation, and whether other pages link to it. An article with real traffic, even modest, is usually worth keeping and differentiating rather than removing outright. Deletion without a 301 redirect to a replacement page simply loses whatever ranking signal and incoming links the page had built up.
Should I combine similar articles?
Where two articles genuinely answer the same question with no meaningful difference in angle, combining them into one stronger page is usually the right call — one substantial, well-structured answer outranks two thin, overlapping ones. Where they answer related but distinct questions, combining them tends to create a page trying to do two jobs at once, which is its own version of the same problem. The test is the same one used for service pages: would a single searcher be equally served by either page? If yes, combine. If no, keep both but sharpen the distinction and link between them.
Can too many pages hurt SEO?
Not directly through volume alone — Google's documentation does not state a page-count penalty. What a large number of pages can do is increase the odds of accidental overlap, dilute crawl and editorial attention, and make it harder for the business itself to know which page is meant to be authoritative for a given question. The problem is rarely 'too many pages'; it is usually 'too many pages trying to answer the same question without anyone noticing'.
Consolidate, differentiate or redirect: deciding which
| Situation | Recommended action |
|---|---|
| Two pages say almost the same thing, same intent, neither performs well | Consolidate into one stronger page; 301-redirect the weaker URL to the survivor |
| Two pages cover related but genuinely distinct angles, both have some traffic | Keep both; rewrite introductions and titles to sharpen the distinction; add internal links between them |
| A service page and an article target the same query | Keep the service page commercial; redirect the article's intent toward supporting it, or use a canonical tag if both must remain for a specific reason |
| An old article is superseded by a newer, better one on the same question | 301-redirect the old URL to the new page so existing links and signals transfer, rather than leaving both live |
| Two pages are near-duplicate variants (e.g. same content for slightly different audiences) and both must exist | Use a canonical tag pointing to the preferred version, per Google's duplicate-content guidance, rather than hoping both rank |
How to check whether this is actually happening
- In Search Console's Performance report, filter by a query you'd expect one page to own and check whether more than one of your own URLs appears for it over time
- Search site:yourdomain.com plus the core phrase and see how many of your own pages Google is willing to surface for it
- Compare the two suspect pages side by side and ask honestly whether a reader would notice the difference in the first two paragraphs
- Check whether each page has distinct, deliberate internal links pointing to it with different anchor text, or whether they are linked identically from the same places
Mistakes to avoid
- Deleting a page with real traffic or backlinks without a 301 redirect, losing accumulated signal for nothing
- Treating every topically similar pair of pages as cannibalisation when they actually serve different buyer stages
- Merging pages without updating internal links that pointed to the removed URL
- Publishing a new article on a question already answered elsewhere on the site, by accident, because nobody checked first
Sources
- Consolidating duplicate URLs — Google Search Central
- 301 redirects — Google Search Central
- Search Console Performance report — Google
- SEO starter guide — Google Search Central
Not sure why the website isn't being found or isn't producing enquiries?
A Commercial Website Audit reviews search visibility, structure, AI readiness and conversion together, and gives you a prioritised 90-day plan. £595 + VAT, fixed fee.
Related services