Forecast or Strategy? Two Very Different Reasons to Model

Earlier this week I was in Boston, where I had a full-day project modeling workshop with a portfolio management team. The training went well, and we had many new insights during the day.

One takeaway that stuck with me is that when we model, we do it for one of two distinct purposes; It’s either forecasting to predict the future, or to gain some insight about our projects or portfolios.

1️⃣When we are forecasting to predict the future, we want to understand what may happen in the project or portfolio. We want to answer questions such as what is the NPV of this project, or how likely is it that we meet our revenue or number of launches target. We set up the models with our best knowledge about the projects and the world they exist in. When we get our outcomes, we are generally pretty happy about that. We can now share the forecast as the agreed view on the future.

2️⃣When we want to gain insights, we want to test strategies, and we generally enter a what-if mode. The goal in this situation is not necessarily to find the “one source of truth” about the project or portfolio, but rather to find directional outcomes and compare these outcomes with a different strategy. On the project level, this could be to test clinical development plans that are loosely defined, and on the portfolio level, it could be to evaluate investment strategies or to evaluate project sourcing strategies.

Forecasts give us a shared reference point, but strategy work demands flexibility, rapid iteration, and a willingness to test bold ideas. Having both capabilities is what makes decision-making truly data-driven.

Thanks again to everyone who joined the workshop and contributed to the rich discussion. If your team would like to sharpen its forecasting skills, stress-test strategic options, and see how a structured modeling framework can boost confidence in tough portfolio calls, reach out—let’s set up a session.