Forecast accuracy is not the objective of a launch team. It is a precondition, and teams routinely mistake the two.
The objective is knowing which decisions change when the numbers shift, and then acting on them. A forecast nobody acts on is a spreadsheet. It holds potential and produces nothing, because no decision was ever wired to it.
What accuracy is actually good for
Accuracy matters. A launch team that cannot forecast within a usable range has no basis for planning supply, sizing a field organization or setting a price. None of that is in dispute, and the analytical effort that goes into precision is not wasted.
What accuracy cannot do is tell you what to do differently. It describes a number. It does not describe a choice. Teams that treat precision as the finish line tend to arrive at a defensible figure and no decision attached to it.
The question worth asking instead
Which decisions change when the numbers shift?
That question forces a forecast to name its own consequences. If demand comes in twenty percent below plan, what happens to pricing strategy, to inventory, to how the field organization is deployed? If the answer is nothing, the forecast was never governing anything. It was reporting.
What a forecast should be changing
A forecast that is doing its job pulls specific conversations into the room. Market access strategy, because coverage assumptions are usually the largest single driver. Patient access, because the gap between a written prescription and a started therapy is where most volume assumptions quietly fail. And revenue impact, stated plainly enough that somebody has to respond to it.
Tied to decisions, a forecast becomes an instrument. Left untied, it is a static report that gets circulated, admired for its rigor and ignored.
What the best launch teams do differently
They treat forecasting as a means rather than an end. They build a habit of adjusting strategy when the data moves, rather than defending the original number until the quarter closes. That turns forecasting from a recurring task into something closer to organizational reflex.
The goal was never to predict the number. It was to be able to respond to the market the number is describing, early enough that responding is still cheap.
The launch version of this argument is in A Launch Does Not Fail in the Forecast. The ten places a patient can leave are on the philosophy page, and terms used here are defined in the glossary. The patient-side view of the same ten exits is at Where did your treatment stop?
The views expressed are my own and are not made on behalf of any current or former employer. All trademarks belong to their respective owners.
Free by email
The 5 Layers of GTN™, in your inbox
Gross-to-Net from someone who has run it. Tuesdays and Thursdays.
Subscribe on Substack