By Sachin Purwar, Partner
The phrase “forecasts are always wrong” is a common refrain in the biopharma industry—and for good reason. Over half of pre-commercial assets fail to meet their projected forecasts, yet forecasting remains an indispensable tool for strategic planning, investment decisions, and resource allocation. A well-constructed forecast considers product attributes, company goals, and market dynamics to guide strategic decisions and tradeoffs and set up the product for commercial success.
Creating a forecast is as much an art as it is a science. With countless variables and evolving market landscapes, it’s no surprise that errors can creep in, sometimes significantly impacting projected revenues. These errors often undermine confidence in forecasts and make it harder to defend them to stakeholders, such as investors, strategic partners, and board members.
Here are five major challenges in creating biopharma forecasts—and strategies to overcome them.
1. Aligning Market Size with the Addressable Population
The starting point for any forecast is understanding the relevant market size, which represents the pool of patients seeking treatment. Often derived from epidemiological (epi) data, market sizing can easily be overestimated if the forecaster fails to consider key nuances. For example:
- Varying patient definitions: Patients might be categorized in a multitude of ways, such as undiagnosed, diagnosed but not seeking treatment, or diagnosed and actively seeking treatment, depending on the epi source.
- Complex disease profiles: Chronic conditions like diabetes or fibromyalgia may involve intermittent symptoms or diagnostic challenges, leading to high variability in estimates.
How to Avoid This:
- Triangulate data sources: Compare multiple epi data sources, analyze their methodologies, and cross-check with real-world data such as claims, sales volumes, or primary research to better define which patient sets are the most relevant for a given asset.
- Clarify assumptions: Clearly document and compare the rationale behind various market size calculations to ensure transparency and adaptability based on hypothesized future market conditions.
2. Avoiding Overcomplication with Too Many Variables
Forecasts often involve dozens of variables, but including too many in an attempt to create a higher degree of accuracy can lead to unnecessary complexity and even double counting. For example, an uptake curve derived from analogs already factors in variables, such as speed of gaining access or reach and frequency, so including these as separate variables for the forecast will lead to redundant downside adjustments.
How to Avoid This:
- Simplify the model: Focus on the most critical variables for a given asset and therapeutic area and ensure clear understanding of how each impacts the forecast.
- Understand analogs: When using analogs for specific forecast inputs, analyze which variables and assumptions are already embedded in their data to avoid overlaps.
3. Managing the Compounding Effect of Variability
Even small errors in input variables can snowball into significant deviations in forecasted revenue, especially with the number of variables involved in a given forecast. For instance, overestimating just two variables, such as the addressable population and uptake, by 20% results in a compounded overestimation of 44%.
How to Avoid This:
- Limit extreme assumptions: When trends or uncertainties warrant aggressive or conservative approaches, apply them sparingly to maintain overall balance. Otherwise, create separate upside or downside cases to capture these uncertainties.
4. Sourcing Reliable Inputs for Variables
Forecasting is a collaborative process, often involving multiple stakeholders with differing perspectives around the future market. Without clear documentation of input sources and methodologies, it becomes increasingly challenging to align teams, gain buy-in, or compare new forecasts to prior ones.
How to Avoid This:
- Ensure transparency: Clearly source and document inputs for every forecast variable, detailing their origins, how they were calculated, and rationale for why they are being considered.
- Gain buy-in: Gaining input across internal stakeholders and understanding different perspectives and nuances about the market can help rationalize the preference of one source over another and build overall consensus.
5. Structuring the Forecast Approach for Market Realities
Every therapeutic area (TA) or indication has unique characteristics and specific country/region nuances. A poorly structured forecasting approach and subsequent model can lead to inaccurate insights and ultimately hamper strategic decisions.
How to Avoid This:
- Conduct thorough market assessments: Gain understanding of all the relevant market variables that can impact utilization, such as the prescriber base, care settings, pipeline, and access channels for the target TA, to help build the hypothesis around the forecast approach
- Leverage insights: Leverage learnings (from primary and secondary research) around patient, prescriber, and payer segments and pressure test how the target asset will play across these stakeholders. This can help fine tune the assumptions and crystallize the opportunity within the context of the markets’ structure and dynamics.
Improving Forecasting to Drive Insights
While no forecast can ever be perfect, taking steps to address these five common pitfalls can significantly improve accuracy, confidence, and credibility. By aligning market size with reality, managing variability, simplifying models, sourcing inputs transparently, and structuring forecasts appropriately, biopharma companies can build forecasts that better withstand scrutiny and drive actionable insights.
Forecasting is not just about numbers—it’s about understanding the complex interplay of variables and telling a defensible story. With the right approach, forecasts can serve as powerful tools for decision making and strategic success.
Learn more about how The Dedham Group’s expertise can help you navigate forecasting challenges. Contact us.
