Effective forecasts should inform many decisions, from earnings guidance to brand-level investment prioritization. To do that effectively, forecasters should be a strategic partner to the broader organization by supporting investment decisions and risk management. Achieving that partnership requires a forecasting organization that’s objective and data-driven, comprehensive and consistent across the portfolio and globally, constantly adapting to a changing healthcare and technology landscape, and focused on enabling real-time organizational decision-making and risk management.


Our research suggests that forecasting leaders today are focusing on changing their process, integrating more data and adopting predictive tools to adapt to these changes. Independently, each of these advances can help increase forecasting efficiency and accuracy. The most effective transformations come from a vision that’s realized through the evolution of many capabilities, not just one or two. Critical to achieving this is making structure, process and technology choices that achieve the goals stated above in an efficient manner.


In our experience, there are three major challenges that forecasting organizations are addressing today: The first challenge is the reconciliation of the variety of forward-looking analytics that are produced on a frequent basis. Internally, forecasting leaders are most concerned about coordinating forecasts globally and across affiliates. Prior to reconciliation, we have seen many issues across organizations, including:

  • Errors in fundamental assumptions across affiliates that don’t have the ability to deeply build all forecasts
  • Thousands of hours wasted redoing similar analyses and presentations across affiliates and functions

The second challenge is incorporating innovation into the forecasting process. Forecasters have lamented that with the current approaches, there’s an abundance of data within an organization, but a limited portion of that data is directly used to inform a forecast. Many companies also have a data science team focused on trying advanced techniques to solve other organizational issues. Some of those techniques, including machine learning, will be valuable for improving the forecasting process when used to supplement judgment and a transparent process rather than replace it. Adapting to analytical innovation will require changes to the typical approach and structure of the forecast models. The flexibility that Excel provides has made it the standard tool for most forecasts, but it’s failing to keep up. Organizations don’t have to abandon Excel entirely (although, in some cases, an alternative approach may reduce error),but they do need to adopt additional tools to support the forecasting process.


Finally, the third challenge will be to use forecasting information and techniques to capture the impact of rapidly changing new dynamics, which will need to be captured in a new forecasting approach. These dynamics include:

  • Gross-to-net (market access): Historically, forecasters have used static assumptions provided by others to include gross-to-nets. Tighter collaboration with value and access teams and more focus on capturing payer/channel contract details and events will be a critical part of accurate forecasting.
  • Indication decomposition: In cases such as immunotherapy in oncology, autoimmune disorders or PCSK9i underlying conditions, forecasting techniques may help companies better understand performance by indication and future forecasting. This issue is especially prevalent in markets outside the U.S. with lower data availability.
  • Pace-of-trial readout: The increasing volume of competitive trials and the focus on specific patient segments or lines of therapy result in granular assumptions about impact and adoption across patient groups.

The three challenges that we’ve identified—overlapping effort, integration of new techniques and increasing complexity—are compelling forecasting teams to rethink their process, skill set and tools in order to provide strategic insight for organizational decision-making. In our upcoming posts, we’ll explore three comprehensive changes that will allow forecasters to increase the value that they provide to organizations.