Key assumptions are the foundation of any model. Modelers should focus the majority of their effort compiling and validating key assumptions. In addition, assumptions need to be clearly communicated so users know the framework the model is built around.
In my experience, there are three primary ways to compile and validate key assumptions:
Use historical data - if your company has a program pipeline, using past data for future results is a good start. These data points can be easily validated through signed contracts and your historical financials. Keep in mind any unique circumstances with the program that you are using the data from. Also, if you are a startup, you likely have no or limited historical data to as a baseline.
Leverage past experience - if you have limited historical data, use your team's experience to build your assumptions. Chances are, someone on your team has been there before. For instance, if you are starting your first clinical trial, your clinical VP likely has cost assumptions from their previous company that they are comfortable with. Be careful though - some people are more budget savvy than others.
Use objective sources - if you are a startup with limited experience and a small team, using a third party resource is a great option. I highly recommend The Pharmagellan Guide to Biotech Forecasting and Valuation as a guide. This book provides figures and metrics applicable for any indication type. I've had great experience using this resource to pressure test my inputs compiled from the sources noted above.
You will likely use some combination of all three of these resources to put together a reasonable set of assumptions. The exercise of compiling and reviewing these assumptions will reveal common and unique circumstances of your programs that need to be accounted for. Reviewing the data from different angles will also ensure nothing is missed.
intheBlk's Long Range Planning tool has a cost assumption interface built to quickly review model assumptions by program, department and development phase. By providing this overview, you can see the full development plan to ensure you didn't miss anything. To find out how intheBlk can help you compile and review model assumptions, contact us below.