Keyword Cannibalization Detector

Find queries where more than one of your pages shows up in search results, see how clicks and impressions split between them, and get a suggested primary page for each.

Use the tool

Why this tool exists

Export the Performance report with both Queries and Pages dimensions (or build a CSV with query, page, clicks, impressions and optionally position). The tool groups rows by query, ignores pages below a minimum number of impressions and lists the competing URLs ordered by clicks.

Several URLs for one query is a prompt to investigate, not always a problem: a guide and a category page can both legitimately rank. Look for cases where the split is close, positions swap week to week, or a weak page outranks the one you want. Then merge, canonicalise, redirect or give each page a distinct intent.

No data is sent anywhere: the work happens in your browser, so you can use it offline once the page has loaded.

How to use it

  1. In Search Console, export queries with pages (or use the API) as CSV.
  2. Paste the CSV; the header needs query, page and impressions columns.
  3. Set the minimum impressions to ignore noise.
  4. Review each query and decide whether to consolidate or differentiate.

Frequently asked questions

Does cannibalization always hurt?

No. Two different result types for one query can both help you. It hurts when the pages are near-duplicates, swap places in rankings, or split links and signals.

What should I do about a conflict?

Merge the pages and 301 the weaker one, point a canonical at the primary page, or rewrite the weaker page for a different intent and update internal links.

Which columns are required?

query (or keyword), page (or url) and impressions. Clicks and position make the report more useful.

Is my data sent anywhere?

No. Everything is processed in your browser, and the page makes no network requests with your input.