Ideas to Action:

Independent research for global prosperity

Taking Forecasting Seriously

Demand forecasting must become imbedded in all global efforts to increase access to essential medicines and technologies. This requires:

  • A clear understanding of what is meant by "demand forecasting" and how it differs from estimating needs and from advocacy and demand creation activities.
  • Investing in technical forecasting capacity and creating models specific to forecasting for developing country health products.
  • Universal adoption of basic principles for good forecasting to increase market understanding and credibility, better understand and mitigate system-wide risk, and increase value for money.

Forecasting Principles

The Working Group recommends 10 basic principles in three categories:

Customer-focused principles ensure that forecasts will meet the needs of customers and have the greatest impact on the decisions they are intended to inform.

1. Identify the principal customers or decisionmakers of the forecast and clearly understand their needs.

2. Understand and clearly communicate the purpose of the forecast and the decisions that it will affect.

3. Create a forecasting process that is independent of planning and target setting.

4. Protect the forecasting process from political interference and ensure it is transparent.

Process- and context-focused principles create a credible forecasting process and help develop, present, and understand the forecast in relation to the overall market and public policy environment.

5. Embed the forecast into the broader environment taking into account market conditions, public policy, competitive forces, regulatory changes, health program guidelines, and the like.

6. Create a dynamic forecasting process that continually incorporates and reflects changes in the market, public policy and health program capabilities.

Methodology- and data-focused principles select the right methods for the nature of the forecast being developed and effectively incorporate qualitative and quantitative information.

7. Choose the methodologies most appropriate to the data and market environment and obtain customers' and decisionmakers' agreement on the methodologies.

8. Keep the methodologies simple and appropriate to the situation, but include enough detail to address the level of investment risk and accuracy required.

9. Make forecast assumptions clear and explicit.

10. Understand data and their limitations, using creativity and intelligence in gathering and introducing data into forecasts.