Audience Overlap: The Hidden Cost Nobody Calculates

A client comes in with a segmentation study. They want five audience segments, n=200 each. Total sample: 1,000. Budget accordingly. Simple, right?

Except three of those segments overlap by 40%. "Working moms 30-45" has massive overlap with "household income $75K+" and "college-educated women." If you don't account for the overlap, you're either going to under-buy sample (because you're assuming 1,000 unique people when you really need 700) or you're going to over-buy (because you think you need 1,000 unique but you only needed 700 with proper allocation).

Either way, you're wrong, and the budget is wrong, and the timeline is wrong.

How the mapper works

Define up to five audience segments with their demographic criteria. The tool calculates the overlap between every pair and renders an interactive Venn diagram showing the unique and shared populations. Each region of the Venn has an estimated population count and feasibility score.

The practical output is a corrected total sample number. If your five segments at n=200 each would be 1,000 unique respondents with zero overlap, but the actual overlap reduces that to 720 unique respondents needed, you just saved the project 28% on sample cost. Or more importantly, you just gave the client an accurate number instead of a guess.

Tech stack

React 18.2 + Babel CDN. The Venn diagrams are rendered in SVG with calculated area proportions based on set intersection sizes. Census ACS data drives the population estimates for each segment. The overlap calculations use multiplicative probability for independent criteria, which is the same simplifying assumption used throughout the feasibility toolkit.