How scattered a query is across your pages
Cannibalization Entropy
Measures how evenly a query spreads across the pages that rank for it, so genuine cannibalization separates from ordinary overlap.
How it works
For each query, the share of impressions each ranking page takes becomes a distribution, and Shannon entropy measures how concentrated it is. A query answered by one page has entropy near zero. A query spread evenly over four pages scores high, which is the shape of real cannibalization. Counting pages alone cannot tell those apart: two pages at 95/5 are not competing, and two at 50/50 are.
What you get
- Entropy per query, with the pages that share it and their impression split
- A threshold you set, so the report matches how strict you want to be
- Queries where the top page changed during the period, a common cannibalization tell
- The pages that appear in the most contested queries, ranked
What it needs
Query and page dimensions for the same period; 28 days or longer.
What it will not tell you
High entropy on a broad head term is often correct rather than a problem. Read it alongside intent before merging or redirecting anything.
Runs on demand
Experimental analyses are not part of the nightly job. You open the report, set the period and run it, so the cost lands only when you want the answer.
The rest of Experimental Analytics
Ready to see what your Search Console data is hiding?
Connect a property and every report fills in immediately. No credit card needed.