Verified-location subset as targeting filter

Restricting delivery to a curated subset of location-verified IPs while measuring against a full-population outcome (store sales) creates a denominator mismatch that dilutes measured lift by an unknowable share — and the subset is not neutral: location-stable IPs skew fixed-line, single-family, owner-occupied, deleting exactly the dense urban geography where fine-grain targeting has the most advantage over ZIP.

Detection

Delivery filtered on a location-quality attribute while the outcome metric counts customers regardless of that attribute.

Mitigation

Use the verified subset as a pre-registered analysis stratum, never as a targeting filter: deliver to everyone, segment delivery logs at readout
Report headline (full population) and stratum (verified subset) side by side; divergence is itself a finding
Run the stability filter at the platform or in a clean room — the measurement party should receive aggregate cell counts by stability class, never raw IPs

Detection

Mitigation

Affected conversions