How to Measure ROI on HCP Marketing Spend

Most HCP programs report something that looks like a return. A measurement partner sends a deck that says "attributed TRx: 30,000" and someone multiplies by price. The number is large and everyone is pleased until finance asks how much of that would have happened anyway. This article sets out the difference between the three numbers that get called ROI, the inputs finance actually needs, and a worked example. It belongs to the HCP engagement strategy series.

ROI, ROAS, and incremental Rx value are different numbers

The step that matters most is the move from attributed to incremental. If the brand targeted its top three deciles, exposed HCPs already wrote more than average before the campaign. Their scripts during the campaign are mostly baseline. The attributed vs. incremental prescriptions piece covers why that gap is often several times the incremental number.

The inputs finance needs

When I have seen ROI estimates rejected, it was rarely because of the media math. It was because one of these inputs was missing or made up.

1. Incremental volume with a range. A point estimate of 9,000 TRx means little without a confidence interval. A range of 5,000 to 13,000 is more honest and helps finance plan.
2. Net revenue per script. Not WAC or list price. Net after rebates, chargebacks, copay support, and other gross-to-net deductions. Brand finance owns this number; ask them for it rather than guessing.
3. Contribution margin. Cost of goods and variable distribution costs come off before you call it return.
4. Full program cost. Working media plus data licenses, identity and match fees, measurement fees, agency fees, creative production, and MLR resourcing if it is significant. The working vs. nonworking costs article lists what usually gets left out.
5. The value window. How many months of refills count. Agree this before launch.

A worked hypothetical ROI calculation

All numbers below are hypothetical and chosen to make the arithmetic easy to follow.

• Working media: $1,000,000. Data, measurement, and agency fees: $200,000. Total cost: $1,200,000.
• Measurement partner reports 30,000 attributed TRx among exposed HCPs over a nine-month window.
• A holdout test estimates 9,000 incremental TRx (range 5,000 to 13,000).
• Net revenue per TRx from brand finance: $400.
• Contribution margin: 85%.

Step by step:

1. Incremental net revenue: 9,000 x $400 = $3,600,000.
2. Incremental contribution: $3,600,000 x 0.85 = $3,060,000.
3. Net return: $3,060,000 minus $1,200,000 = $1,860,000.
4. ROI: $1,860,000 / $1,200,000 = 1.55, or 155%.

Now look at what the inflated versions would say. Attributed TRx at net price: 30,000 x $400 = $12,000,000, a "10x return" on $1.2M. Use list price of a hypothetical $700 and it becomes $21,000,000. Both numbers are arithmetically correct and commercially meaningless. The honest answer is 155%, with the low end of the range (5,000 TRx) giving roughly 42% and the high end (13,000 TRx) roughly 268%.

Check the low end: 5,000 x $400 x 0.85 = $1,700,000; minus $1,200,000 is $500,000; divided by $1,200,000 is about 0.42. That range is the real finding. It says the program very likely paid back, and it tells finance how much uncertainty to plan around.

Common ways HCP ROI gets inflated

• No comparison group. Exposed vs. unexposed without matching on prior volume and tier just measures targeting.
• List price instead of net. For many brands the gap between list and net is large.
• Media cost only. Leaving out data, measurement, and fees can shrink the denominator by a fifth or more.
• Overlapping credit. Field, email, and media each claim the same incremental script. Sum the channel ROIs and you get more incremental volume than the brand actually grew.
• Window shopping. Choosing the attribution window after seeing the results. Set it in the measurement plan; the guide to choosing an Rx attribution window explains how.
• Counting lifetime value without saying so. Projecting a new patient's refills over two years can be reasonable, but label it as a projection with its own assumptions.

When you cannot measure incrementality

Some programs are too small, too short, or too entangled with field activity for a clean holdout. In that case, say so. Report engagement state movement and attributed volume as diagnostics, and estimate a break-even instead: how many incremental scripts would the program need to pay back? In the example above, break-even is $1,200,000 / ($400 x 0.85), about 3,530 TRx. A brand lead can then judge whether that is plausible given reach and history. That is a more credible conversation than a precise ROI built on attribution alone. For larger portfolios, marketing mix modeling can estimate channel contribution where NPI-level tests are not possible.

Practical takeaway

Before your next measurement readout, calculate the break-even incremental TRx for the program using net revenue, margin, and full cost. Put it on the first slide next to the incremental estimate and its range. If the low end of the range is above break-even, you have a strong case. If break-even sits inside the range, you have an honest one, and you can decide together what more evidence is worth paying for.

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.