Keyword cannibalization is what happens when two or more of your own pages target the same search query. Google picks one — often not the one you would pick — splits the clicks and links between them, and ranks both lower than a single focused page would rank. This guide shows how to find it in Search Console and how to fix it by merging, redirecting, or differentiating. It is the fifth cause in our pillar guide, why your articles aren't ranking.
- Cannibalization is two of your own pages chasing the same query: Google splits the signals, often picks the weaker page, and ranks neither as well as one focused page would rank.
- Find it in Search Console in about ten minutes: one query, two of your URLs, alternating impressions.
- The three fixes are merge, redirect, or sharply differentiate — chosen by which page deserves to own the query.
- It is not a penalty. Nothing is punished; the signals are simply divided.
What it looks like in the wild
- A query in Search Console shows impressions for two different URLs of yours, alternating over time.
- Your "main" article for a topic keeps swapping with an older or thinner one in the results.
- A page's position for its key query bounces between page 1 and page 3 week to week with no changes on your end.
- You published an "updated" version of a post and left the old one live at a similar URL.
Finding it in Search Console (10 minutes)
- Open Performance → Search results. Set the date range to the last 3 months.
- Click the Queries tab, pick an important query, then switch to the Pages tab. If two or more of your pages earned impressions for that query, note them.
- Repeat for your top 10–20 queries — cannibalization usually clusters around your most valuable topics.
- For each pair found, open both pages and answer honestly: are these trying to satisfy the same searcher with the same job? If yes, it is cannibalization, not "covering a topic from two angles".
Not every multi-page query is a problem: a hub and its spoke can legitimately both appear if they serve clearly different intents (overview vs. one narrow sub-question). The test is sameness of intent, not sameness of topic.
The three fixes
1. Merge (most common, most effective)
Combine the pages into the stronger URL: fold the weaker page's unique material into it, redirect the weaker URL to the survivor (301), and update internal links. Choose the survivor by backlinks, traffic history, and URL quality — not by which one you wrote most recently. Merged pages routinely outrank both parents because every signal now points at one target.
2. Differentiate (when both topics are real)
If the two pages could serve genuinely different intents, make that difference unmistakable: retitle both for their distinct queries, rewrite introductions to state each page's specific job, reassign internal links so each page is linked with anchors matching its own target, and remove overlapping sections. Then re-check in a month — if they still alternate for the same query, merge instead.
3. Demote one (hub vs. spoke discipline)
Sometimes the fix is hierarchy: the spoke is outranking the hub for the broad query because the hub is weak. Strengthen the hub (depth, freshness, internal links) and make sure the spoke's title and anchors target its narrow query only. The cluster model in hub-and-spoke content clusters is designed to prevent exactly this collision.
Prevention: a one-line publishing rule
Before writing anything, search your own site for the topic. If a page exists that targets the same intent, your options are update it, expand it, or write a deliberately different-intent page — never publish a second version and hope. Our pre-publish checklist includes this check as step one, and the refresh guide covers the update-instead-of-duplicate habit.
A note on the word "penalty"
Google representatives have repeatedly said there is no cannibalization penalty — and they are right in the narrow sense: nobody is punished. But the ranking outcome is the same as a small penalty: split signals, unstable positions, and the weaker page shown. The fix is editorial, not a reconsideration request.
Sources (5)
Google Search Console Performance report documentation (support.google.com/webmasters); Google Search Central blog discussions of duplicate content and canonicalization (developers.google.com/search/blog).
- Google Search Central — How to specify a canonical URL
- Search Console Help — URL Inspection tool
- Google Search Central — What is a sitemap
- Google Search Central — Google Search Essentials
- Google Search Central — SEO Starter Guide
Worked example (the classic blog pattern)
Consider a cooking blog with "best cast iron skillet" (2023 roundup, 40 backlinks, position slipping) and "cast iron skillets: the complete guide" (2025, better written, no links). Both earn impressions for the same queries; the 2023 post shows most weeks because its links outweigh the 2025 post's quality. The wrong fixes: rewriting the 2023 post again (third competitor enters), or deleting the 2025 post (losing the better content). The right fix: merge — move the 2025 guide's superior sections into whichever URL has the stronger profile (here, likely the 2023 URL after updating its content wholesale, or the 2025 URL with a 301 from 2023 so its links transfer to the better page). Publishers flinch at redirecting the linked page; but a 301 passes the link value to the survivor, which is the entire point. Six weeks later there is one page, holding both the content quality and the link equity, sitting where neither parent could.
What about deliberate multi-page coverage?
Brands sometimes rank two results for one query and call it strategy. It happens legitimately when the pages serve different formats (a guide and a video, a review and a comparison). If both your pages are text articles with the same intent, though, "strategy" is usually a story told about an accident. Apply the intent test ruthlessly; genuine format differences pass it, hopeful thinking does not.
After the fix: consolidation etiquette
Once you merge or differentiate, tidy the blast radius. Update every internal link that pointed at the retired URL to point at the survivor directly (do not rely on the redirect as a permanent internal signpost). Check that the survivor's title, intro, and hub listing describe its now-broader job. Then leave the pair alone for a full month: the temptation to fiddle during consolidation is strong and counterproductive, because each new change restarts your uncertainty about which change worked. Log the baseline, the action, and the review date — the same discipline as a content refresh.
Record-keeping for merges
Keep a merges tab beside your refresh log: retired URL, survivor URL, date, the query in dispute, and the baseline positions of both pages. Six months later, that tab answers questions memory cannot — whether merging is a reliable lever on your site, and which survivor-choice rule (links vs. content quality) predicted the winner.
Frequently asked questions
How many pages can I have on one topic?
As many as serve genuinely different search intents. A pillar plus five narrow spokes is healthy; two near-identical 'ultimate guides' to the same query is cannibalization.
Should I delete the weaker page instead of redirecting?
Redirect (301) in almost every case: it preserves any links and history the page earned. Delete only pages with no links, no traffic, and content fully absorbed elsewhere — and even then, redirecting is usually safer.
How long until rankings settle after a merge?
Typically 2–6 weeks for the surviving page to consolidate positions. Keep internal links consistent during that period and avoid further changes to the pair.
Can cannibalization happen between a post and a category page?
Yes — tag and category archives that mirror a guide's topic and title are a classic example. Usually the fix is noindexing thin archives or retitling them to be clearly navigational.
What if both pages rank well, one at #4 and one at #6?
That looks like success but usually means one merged page could sit at #1–2 and take most of both pages' clicks. Test carefully on your most valuable queries first.
Run the Search Console check for your main queries. Where two of your pages are splitting one query, merge them or differentiate them sharply — and let a single strong page collect the signals both were starving for.
Did you find a query where two of your pages alternate in the results — and which one should own it?
