Keyword cannibalization checker: find pages competing for the same query

What keyword cannibalization means

Keyword cannibalization is when two or more pages on your own website compete for the same search query, so Google splits its attention between them and neither page ranks as well as one clear page would. You see it as impressions for one query spread across several of your URLs, or as the ranking URL changing from week to week while the position stays stuck.

It usually happens by accident. A blog post and a service page both target the same phrase. An old guide and its newer update both stay live. A category page and a product page share a title. Each page alone might be fine; together they send mixed signals about which one you want Google to show.

Two URLs appearing for one query is not automatically a problem. Brand searches, one page that already dominates, and pages that serve genuinely different intents are usually healthy. The checker is built to separate those cases from real overlap.

What the checker uses as evidence

The checker works from your own Google Search Console data, not from scraped results or third-party estimates. For each query it needs the pages that appeared, with their clicks, impressions and average position over the reporting window.

• Query and page together. Each row is one query–page pair. Search Console's separate Queries and Pages tables cannot be joined reliably, because neither says which page earned which query.
• Clicks and impressions. These show how much real demand each page is getting for the query, and how the total is split.
• Average position. This is Search Console's average over the window, not a live rank at one moment. It tells you whether the competing pages are close enough to the top to matter.
• Your site URL. Used only to recognise and skip brand queries, so a search for your own name is not treated as cannibalization.

Search Console reports /page/#section rows separately from /page/. The checker folds those fragments into one page before comparing, because they are the same URL.

How the checker decides what counts

Not every overlap is worth your time, so the checker applies the same rules as MrKeyword's in-app analysis. A query is listed only when all of these are true:

1. At least two pages genuinely compete

   Each competing page must have 5 or more impressions, at least 10% of the query's impressions, and an average position of 20 or better. A page that appeared once at position 60 is not competing.

2. No single page already dominates

   If one page holds 80% or more of the query's impressions, the query is treated as healthy and left out. There is nothing meaningful to consolidate.

3. Brand queries and search operators are skipped

   Queries containing your domain label are skipped, because a second brand result is extra space on the results page, not competition. Search operators such as site: are skipped too, because every page matches them by design.

4. Duplicate rows are merged first

   Fragment URLs are folded into their page, and positions across duplicate rows are combined as an impression-weighted average, so one page is not counted twice.

How to interpret overlapping URLs

Each result is a potential conflict that needs an intent review, not proof that you should redirect a page. Read the competing URLs and ask what each one is for.

Positions and shares are Search Console averages for the reporting window. They are evidence for a decision, not a forecast of what will happen after a change.

A concise example workflow

Here is how one query moves from a flagged overlap to a finished decision.

1. Connect Search Console and open the checker

   MrKeyword syncs your query-and-page rows. The checker lists the queries with the most shared impressions first, so the biggest overlaps are at the top.

2. Read the competing pages

   For one query, note each URL's share and position. Open both pages and compare their titles, headings and the job each one does for the reader.

3. Choose the page to keep

   Pick the page that best answers the query, or the one with the stronger position and more impressions. If the pages serve different intents, keep both and make them clearly different.

4. Make the change on your site

   Merge useful content into the page you keep, then redirect the retired URL or add a canonical if it must stay live. Update internal links to point at the kept page directly.

5. Check again after a few weeks

   Compare the query's Pages tab four to eight weeks later. The goal is one URL holding most of the impressions, at a better average position.

Practical next steps

• Start with the queries that have the most shared impressions. They are the ones where a clear decision is most likely to matter.
• Decide intent before you touch URLs. A redirect is hard to undo and can remove a page that was earning its own traffic.
• Use a permanent redirect when you retire a duplicate, and a rel="canonical" when the duplicate must stay accessible. Google treats both as strong signals.
• Point internal links at the page you keep, with descriptive anchor text, so Google and visitors reach the right page directly.
• Record the date and the change. Then judge the result from Search Console data, not from a single day's position.

For the full decision table and Google's guidance on redirects and canonicals, read how to find and fix keyword cannibalization. To review every keyword you already rank for and the page behind it, use the step-by-step keyword audit. The same detector feeds the SEO opportunities inbox, where competing pages appear as a ranked fix with a suggested page to keep.

What the checker does not do

• It does not merge pages, add redirects or edit canonicals for you. It shows the measured overlap and suggests a page to keep; you make the change on your site.
• It does not promise a ranking gain. Whether a page moves after a merge or redirect depends on the content, the competition and how Google evaluates the result.
• It only sees what Search Console reports. Queries Google anonymises for privacy are not in any keyword row, so cannibalization on very rare queries cannot be detected.
• It does not judge intent for you. Two pages flagged as competing may serve different needs; read both before merging.
• It does not check other people's sites. Cannibalization detection needs your own connected Search Console property.

Which plans include it

The cannibalization check has no separate allowance: every plan gets it for every website it can hold, and it reads the Search Console data you have already connected.

See pricing for every limit.

Questions and answers