How to Find Keyword Cannibalization (2026 Guide)

What is keyword cannibalization

Keyword cannibalization is when Google can't decide which of your pages to rank for a query and rotates between them. You'll see the same keyword's ranking URL change every few days — one week your blog post is ranking, next week it's your category page, and neither ever quite settles into the top 3.

The cost isn't just position volatility. Every internal competitor splits authority signals: half your backlinks point at page A, half at page B, and neither accumulates enough to beat a competitor page with a single canonical URL.

Symptoms to look for

• The same query returning different URLs on your site across weeks.
• A big drop in CTR without a matching drop in impressions.
• Two internal pages both listed in Search Console for one query, both in positions 6–15.
• A page that ranked well suddenly dropping when you publish a similar new page.

Step-by-step: find cannibalized keywords manually

1. Open Google Search Console → Performance → Queries.
2. Filter to the last 90 days and export as CSV.
3. In a spreadsheet, add a column with the ranking URL for each query row.
4. Group by query and count unique URLs.
5. Any query with 2 or more URLs both ranking in positions 1–20 is a cannibalization candidate.

How LemRank flags cannibalization automatically

LemRank's Cannibalization detector runs on your full rank history — not just Search Console's 1,000-query cap — and looks for two signals in combination:

• URL flip-flops: the same keyword landing on different URLs across at least 3 rank checks in a 14-day window.
• Position clustering: 2+ of your pages appearing in the top 20 for the same query on the same check.

Every flagged case includes a suggested fix — usually one of the four patterns below.

The four ways to fix cannibalization

Pick the fix that matches the intent overlap:

Fix	When to use it
Merge	Two pages cover the same topic. Combine into one, 301 redirect the weaker URL.
De-optimize	You want both pages to survive. Rewrite the weaker page to target a sibling keyword instead.
Internal-link	Pages target related but distinct queries — clarify hierarchy with strong internal links to the winner.
Noindex	One page is a low-value duplicate (tag archive, filter combination) — noindex it.

Common mistakes

• Assuming any two-URL overlap is cannibalization. If both pages consistently rank in different positions (e.g. #2 and #14), it's usually a hierarchy signal, not a conflict. True cannibalization shows flip-flops.
• 301-redirecting without checking backlinks. Always redirect the lower-authority URL to the higher, not the other way round. Losing links because you redirected the wrong direction can drop rankings further.
• Merging pages with different search intent. "How to do X" and "X software" look similar but target different SERPs. Merging them makes both worse.
• Fixing every flagged case in one week. Cannibalization fixes take 2–4 weeks to settle in Google's index. Batch them so you can measure impact.

What this does not do

• It doesn't rewrite your content — it flags the URLs that need attention.
• It doesn't push 301s automatically — you decide which URL wins.
• It doesn't stop cannibalization from happening again — your editorial calendar has to enforce one-page-per-target-keyword going forward.

Related

• How to detect content decay
• How to do a keyword gap analysis