HCP-Level vs. Geographic Pharma Measurement Designs
HCP-level test-and-control vs. geographic experiments for pharma media, compared on spillover, sample size, control quality, and privacy.
The short answer
When a brand wants to measure whether media causes prescriptions, it needs a comparison group. There are two broad ways to create one. HCP-level designs assign individual prescribers to exposed or unexposed groups. Geographic designs assign regions, such as markets or ZIP code clusters. Each has strengths. The choice depends on the channel, the audience size, and how much spillover you expect.
HCP-level designs
Individual prescribers on the target list are randomly split. One group receives media; the other is held out.
Strengths - Many units, which usually means more statistical power. - Direct link between the targeted prescriber and their prescribing. - Works well for NPI-targeted display and email.
Weaknesses - Spillover: held-out prescribers may still see ads through other channels or colleagues. - Not all channels can hold out specific individuals. Broad video and CTV often cannot. - Requires reliable NPI-level delivery control.
Geographic designs
Regions are assigned to test and control. All media in test regions runs; control regions receive none or reduced media.
Strengths - Works for channels that cannot target individuals, such as CTV and broad video. - Captures spillover within a region as part of the effect. - Does not require individual-level identity.
Weaknesses - Fewer units, which means less power and more sensitivity to local events. - Regions differ in many ways; balancing them is hard. - National campaigns and field activity can contaminate control regions.
The design choice matrix
| Factor | Favors HCP-level | Favors geographic |
|---|---|---|
| Channel can target individual NPIs | Yes | |
| Channel is broad or household-based | Yes | |
| Large target list | Yes | |
| Strong expected spillover between prescribers | Yes | |
| Need to avoid individual-level data | Yes | |
| Other brand activity varies by region | Yes |
Handle spillover honestly
Spillover happens when control units are affected by the treatment. In HCP designs, it biases lift estimates down, making media look less effective. In geographic designs, it can occur across region borders. Neither design eliminates spillover. Say how much you expect and whether the estimate is likely conservative.
Plan for power
Estimate how large an effect you need to detect and whether the design can detect it. With few geographic units, only large effects will be detectable. NIST's handbook on confidence intervals is a plain reference for how sample size affects precision. If the design cannot detect the effect that would change your decision, change the design. See holdout test design.
Hybrid approaches
Some brands combine the two: HCP-level holdouts for display and email within geographic test regions for video. This is more complex to run and analyze but can answer channel-specific questions. Get analytical help before attempting it.
Practical takeaway
Choose the design by the channel and the decision, not by what the measurement vendor usually offers. Write down the expected spillover direction and the smallest effect the design can detect before launch.
Sources
External guidance and platform documentation change. Links were current at publication; check them again before relying on them for a decision.
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.
New pharma programmatic breakdowns, occasionally
One email when I publish something worth reading. Benchmarks, measurement teardowns, and case studies with the caveats attached. No cadence promises, no reselling your address.
Unsubscribe any time. See the privacy policy.
Working through this decision on a real plan?
I work on health and pharma data, identity, and activation, after five years running HCP and DTC programmatic agency-side. Happy to talk through how this applies to your situation.