How to Measure Audience Overlap Before Buying More Data

Define the denominator

Choose the population relevant to the decision: eligible HCP list, addressable patients, households, or devices. Do not mix units. A household overlap rate cannot be compared directly with person-level or NPI-level overlap.

For two audiences A and B, calculate:

• overlap count: people in both A and B;
• overlap rate relative to A: intersection divided by A;
• Jaccard similarity: intersection divided by union;
• incremental B reach: B excluding A;
• marginal cost per incremental qualified person.

Use a common test environment

Differences in onboarding, cookies, device graphs, and platform availability can look like data differences. Whenever feasible, onboard vendors into one controlled environment with the same date, geography, suppression logic, and identity provider.

Diagnose overlap quality

High overlap can validate a core population or signal redundancy. Low overlap can indicate useful reach, different construction, or poor precision. Examine composition by specialty, priority tier, geography, age band where appropriate, and addressability. Avoid interpreting sensitive health characteristics beyond approved, privacy-reviewed uses.

Hypothetical example

Vendor A resolves to 100,000 eligible households. Vendor B resolves to 70,000, of which 50,000 overlap A. B therefore contributes 20,000 incremental households before media delivery. If B costs $60,000, the audience cost is $3 per incremental eligible household. This does not establish outcome value; it only frames the reach economics.

Add outcome evidence carefully

Audience quality and business outcomes are separate. A partner can reach an intended population accurately but fail to change behavior. Conversely, apparent outcome strength can reflect baseline differences if groups were not comparable. Use randomized or well-designed comparison methods where feasible and retain uncertainty.

Practical takeaway

Buy the marginal audience, not the headline audience. The next step is a one-page overlap matrix that reports identity unit, eligible union, incremental contribution, composition, cost, and outcome evidence separately.

Editorial note. Analysis and frameworks are the author's own and do not represent Acxiom or any current or former employer, client, or named platform. Examples labeled hypothetical or illustrative are not results from real campaigns. Nothing here is legal, regulatory, or medical advice.