How to Forecast Incremental Prescriptions Without Pretending Certainty

Build the model from auditable components

One illustrative structure is:

incremental eligible people reached × baseline Rx probability × relative lift

Adjustments may be needed for identity coverage, treatment persistence, overlap, ramp, and data maturity, but avoid adding factors that cannot be explained or tested.

Avoid double discounts

If observed historical lift already reflects identity and claims coverage, applying those discounts again understates the result. Conversely, extrapolating a matched-population result to all impressions can overstate it. Write the population attached to every input.

Hypothetical worked example

Assume 200,000 incremental eligible people reached, a 2% baseline prescription probability, and a plausible relative lift of 3% to 8%. The projection is 120 to 320 incremental prescriptions: 200,000 × 0.02 × 0.03 through 200,000 × 0.02 × 0.08. These numbers are illustrative and not campaign results.

That range still depends on design transferability, overlap, exposure quality, and scale. It should not be converted to revenue without finance-approved value and persistence assumptions.

Run sensitivity and break-even views

Identify which input changes the decision most. Solve for the effect required to break even rather than reporting only expected ROI. Show downside, base, and upside values and compare them with the cost of waiting for better evidence.

Reconcile forecast with actuals

After a mature study, update inputs, document error, and retain the old version. Do not rewrite the original forecast. Over time, forecast calibration can become more valuable than a single favorable result.

How to present a forecast to leadership

Forecasts are often read as promises. Presentation choices can prevent that.

Lead with the range, not the midpoint. Say "we project 120 to 320 incremental prescriptions" before mentioning a base case. If you lead with a single number, that number becomes the target.

Name the biggest uncertainty. Tell leadership which input moves the forecast most and what evidence would narrow it.

Show the break-even line. If the low end of the range falls below break-even, say so plainly and explain how a staged plan manages that risk.

Label it clearly. Every slide or dashboard showing forecast figures should say "projection" in the title. Mixing projected and measured numbers on one chart without labels invites misreading.

Commit to a reconciliation date. Tell leadership when you will compare the forecast with measured results. Following through builds credibility for the next forecast. See how to write a media investment thesis for where the forecast fits in a funding request.

Practical takeaway

Forecasts are decision aids, not outcomes. The next step is to publish an assumption register beside every projection and stamp dashboards “projected,” “measured,” or “modeled.”

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.