Keyword cannibalization happens when two or more pages on your site target the same search intent, so they compete with each other in Google instead of with other sites. The usual symptoms are a ranking URL that keeps switching and positions that stall. The fix is to choose one preferred page and merge, re-target or re-link the others.
Key takeaways
- Cannibalization is about intent, not wording: two pages that answer the same need compete even when their keywords differ.
- It only matters when it costs you: a flipping ranking URL, stalled positions or the wrong page ranking.
- Find it in Google Search Console (filter one query, open the Pages tab) and confirm it with a rank tracker that records the ranking URL.
- Merge and 301 same-intent pages, re-target different intents, canonicalize true duplicates and noindex pages with no search value.
- Prevent it with a keyword map that gives every primary keyword exactly one URL.
What is keyword cannibalization?
Keyword cannibalization is when several pages on one site are built to answer the same search, so Google sees more than one candidate from you and has to pick. It is also called SEO cannibalization or content cannibalization (British spelling: cannibalisation). The term comes from marketing, where a new product eats into an existing one's sales.
Search intent decides whether two pages compete, not identical wording. "How to train a puppy to sit" and "teaching your puppy the sit command" have the same intent, so two posts built around them compete. "Puppy training" and "puppy training classes near me" share words but not intent: one wants a guide, the other a local service. Our guide to search intent shows how to read intent from the results page.
Typical causes: old posts never updated or retired, a blog post and a product page optimized for the same term, tag pages that repeat article titles, and writers publishing without a shared keyword plan.
Keyword cannibalization vs duplicate content
Duplicate content is the same or nearly the same text at more than one URL; keyword cannibalization is different pages chasing the same intent. The first is a technical problem that redirects and canonical tags solve, the second a content-planning problem. Neither is a penalty: Google's SEO Starter Guide says duplicate content "is not a violation of our spam policies" and is "not something that will cause a manual action".
Why is SEO cannibalization a problem?
SEO cannibalization is a problem because Google shows only a few pages from one site for a query, so extra pages on the same intent rarely win extra space; they split what one page could have earned. Google's guide to its ranking systems says its site diversity system "generally won't show more than two web page listings from the same site" in the top results, unless more are especially relevant.
The cost shows up as split signals (backlinks and internal links divided between two URLs), an unstable ranking URL (Google alternates between your pages and neither settles), the wrong page ranking (an old post outranks the product page that converts) and wasted work (two pages to update for one topic).
Does keyword cannibalization affect AI search?
Keyword cannibalization can matter for AI search too, because Google's AI features find their sources through search. Google says AI Overviews and AI Mode may use "query fan-out", "issuing multiple related searches across subtopics and data sources", and that no special optimizations are needed to appear in them (Google Search Central). One complete page per intent is a stronger candidate for those related searches than two half-finished ones.
When is keyword cannibalization not a problem?
Keyword cannibalization is not a problem when the overlap costs you no rankings, clicks or conversions. Leave the pages alone when:
- Both pages rank near the top together. Two listings in the top five take space from competitors.
- The intents differ. A guide and a product page can both appear for "standing desk" when Google shows both kinds of page.
- The second page appears rarely and far down. A few impressions at position 40 don't compete with your page at position 4.
- The pages serve different countries or languages, or the query is your brand name.
- The movement is short-term, right after a publish or update.
As a rule of thumb (ours, not Google's), act when the second URL takes a real share of the query's impressions, say a fifth or more, or when the ranking URL changes between checks and the position drops with it.
How do you check for keyword cannibalization?
Check for keyword cannibalization with three tools: Google Search Console for the pages that earn impressions, a site: search for forgotten pages, and a rank tracker for changes in the ranking URL.
Method 1: Google Search Console (query to pages)
- Open Performance > Search results and set the date range to at least three months.
- Filter by the exact query, or by "Queries containing" to include close variants.
- Open the Pages tab. Every URL listed got impressions for that query.
- Compare clicks, impressions and average position per page.
Google's Performance report documentation notes that data grouped by queries is "aggregated by property", while data grouped by pages is "aggregated by page". That is why the Pages tab, with the query filter on, is where you see how one query splits across your URLs. For a site-wide review, repeat this for your top 20 to 50 queries.
Method 2: a site: search
Search site:yourdomain.com "your keyword" to list the indexed pages on your domain about a topic. It is not a ranking report, but it quickly surfaces old posts, tag pages and test pages indexed by accident.
Method 3: watch the ranking URL in a rank tracker
A rank tracker that records the ranking URL shows cannibalization as it happens: the page alternates between two of your URLs and the position zigzags with it, a pattern Search Console's date-range averages hide.
How to do this in RankTera
RankTera's rank table has an Origin column ("Your page that ranks") showing which of your pages ranked for each keyword at the latest check. Origin shows that latest check only, so use Export Table after each daily or weekly check (the Excel file includes an Origin column) and compare the files: a keyword whose Origin alternates between two URLs while its rank bounces is a candidate, and its Rank Progress chart shows how the position moved. The weekly ChatGPT and Gemini check also lists up to 10 cited source URLs per engine with your page highlighted, so you can see which of your pages AI answers cite.
Confirm the problem before you fix it
Fix it when at least two of these are true: the competing pages serve the same intent as the current top results; the second URL takes a real share of impressions or sometimes outranks your preferred page; the ranking URL has switched in recent checks; the page that ranks is not the one you want.
How to fix keyword cannibalization: a decision table
How to fix keyword cannibalization depends on two questions: do the pages serve the same intent, and must both stay live for visitors? First choose the preferred page (usually the one with the most backlinks, conversions and depth), then match your situation:
Situation | Fix | What to do | Watch out for |
|---|---|---|---|
Two pages, same intent, overlapping content | Merge + 301 redirect | Move the useful content into the preferred page, then permanently redirect the other URL | Move unique sections first; keep the redirect |
Pages that should serve different intents but read alike | Re-target | Rewrite the weaker page's title, H1, intro and angle around its own keyword | Confirm the intents differ on the results page |
Both pages fine, but links point to the wrong one | Fix internal links | Point topic anchors and menus at the preferred page | Old posts carry most of the stray links |
Near-identical pages that must both stay live (print, filtered or variant URLs) | rel="canonical" | Add a canonical tag on the duplicate pointing to the preferred URL | For duplicates, not two different articles |
A page visitors need but search doesn't (thin tag archive, internal search results) | noindex | Add a noindex robots meta tag | Don't also block it in robots.txt |
Both pages rank near the top, or intents differ | Do nothing | Keep tracking the ranking URL | Re-check after big changes |
Merge and 301 redirect
Merging is the default fix for same-intent pages: combine the best sections into the preferred page, then redirect the old URL with a permanent 301 or 308. Google's redirects documentation recommends "a permanent server-side redirect whenever possible" when a page moves and says Google uses the redirect as a signal that the target should be canonical.
Re-target the weaker page
Re-targeting keeps both pages but gives each its own job: rewrite the narrower page around a specific long-tail keyword and link it to the broad guide.
Fix internal links
Internal links tell Google which page you consider the main one. Google's link guidelines say anchor text "tells people and Google something about the page you're linking to". If most topic anchors point at the secondary page, repoint them to the preferred one.
Use rel="canonical" and noindex only where they fit
A canonical tag tells Google which of several near-identical URLs to show. Google's canonicalization guide ranks redirects first among the methods and calls rel="canonical" "a strong signal", so use a canonical only when both versions must stay live. A noindex rule drops a page from Google's results entirely, and the canonicalization guide warns: "We don't recommend using noindex to prevent selection of a canonical page within a single site."
After any fix, keep only the preferred URL in your XML sitemap, request indexing for it in Search Console's URL Inspection tool, and watch the ranking URL for a few weeks.
Worked example: two remote-work guides competing
Example with illustrative numbers: an HR software company has two posts: Page A (/blog/remote-team-management, 2023, linked from 11 sites) and Page B (/blog/how-to-manage-remote-teams, 2025, linked from 2 sites). Search Console for "how to manage remote teams", last three months:
Page | Clicks | Impressions | Avg. position |
|---|---|---|---|
Page A | 46 | 5,200 | 12.4 |
Page B | 19 | 3,100 | 15.8 |
/features/remote-teams | 0 | 140 | 48.0 |
Over eight weekly rank checks, the ranking URL switched between A and B five times, and the position moved between 9 and 17.
Diagnosis. A and B are both how-to guides, like the top results; B takes 37% of the impressions and the ranking URL keeps flipping, so this is cannibalization. The features page appears rarely, far down and with a different intent, so it stays.
Fix. The team keeps Page A (more referring sites, older URL), moves B's one unique section on asynchronous meeting rules into A, sets a 301 redirect from B to A, repoints the nine internal links that went to B, removes B from the sitemap and requests indexing for A.
What to watch. Once Google recrawls both URLs, Page A should rank at every check and the position should stop zigzagging; whether it ranks higher still depends on the competing pages.
How to prevent keyword cannibalization with a keyword map
A keyword map prevents keyword cannibalization by giving every primary keyword exactly one URL. It is a simple spreadsheet:
Primary keyword | Intent | URL | Secondary keywords |
|---|---|---|---|
how to manage remote teams | Guide | /blog/remote-team-management | remote team management, managing remote employees |
remote meeting rules | Narrow guide | /blog/remote-meeting-rules | async meeting rules |
remote team software | Commercial | /features/remote-teams | remote work tools for teams |
- List every indexable URL from your XML sitemap or a crawl.
- Give each URL one primary keyword and one intent, with close variants as secondary keywords.
- Check the map before you write; if the intent already has a URL, update that page instead.
- Plan in clusters: a pillar owns the broad term, supporting pages own narrower questions (see our guide to topic clusters).
- Review the map quarterly against Search Console.
For an existing site, RankTera's Scan Website for Keywords reads your sitemap, lets you pick pages and suggests keywords per page to track. New to this? Start with our keyword research guide.
What we found: how the top pages cover keyword cannibalization
We ran "keyword cannibalization" through RankTera's KW Research on October 15, 2026, using US Google results. The pages it analysed (semrush.com, yoast.com, ahrefs.com, smartsites.com and splitmetrics.com) averaged 2,031 words, and the run found 81 keywords. The five most-used terms, plus two that shape the fix, with their average count per page:
Term | Average count per page |
|---|---|
keyword cannibalization | 10 |
multiple pages | 5 |
two pages | 3 |
Google Search Console | 3 |
search engines | 3 |
search intent | 2 |
preferred page | 2 |
Google Search Console is the only tool name among them, and "preferred page" appears about twice per page: the same approach this guide follows, finding the overlap in Search Console and consolidating on one preferred page. Our own analysis of six of the top 10 Google results (US, the same day) found main texts of 956 to 3,532 words, with a median of about 2,175, and Google showed an AI Overview for the query.
Frequently asked questions
Is keyword cannibalization a Google penalty?
No. Keyword cannibalization is not a penalty and does not cause a manual action, just as duplicate content doesn't. Google simply has to choose which of your pages to show for the query. The harm is indirect: split links, an unstable ranking URL and sometimes the wrong page ranking.
Can two pages from the same site rank for the same keyword?
Yes. When two of your pages both rank near the top for one keyword, that is a good outcome, not cannibalization; Google generally allows a site up to two listings in the top results. Act only when the pages take turns ranking, push each other down, or the wrong page ranks.
Should I use a canonical tag or a 301 redirect?
Use a 301 redirect when one page no longer needs to exist: it sends visitors and Google to the preferred page. The primary purpose of a canonical tag is to tell Google which of several near-identical URLs to show while every version stays live. For two different articles on the same topic, merge them and redirect rather than relying on a canonical.
How long does it take to recover from keyword cannibalization?
Recovery depends on how quickly Google recrawls the old and new URLs, which can take days on an active site and longer on a small one. Judge the fix over several weeks: the first sign is a stable ranking URL. No fix guarantees a higher ranking.
Catch keyword cannibalization before it costs you rankings
Keyword cannibalization is easiest to fix before positions slide. RankTera checks your Google positions for every tracked keyword, shows the page that ranks in the Origin column and records whether ChatGPT and Gemini cite you. Start your 14-day free tier — no card needed: it includes weekly tracking for 300 keywords, and the pricing page lists the paid plans.