The further we look into the future, the less certain we become. Consider the following questions: What will you be doing one minute from now? How about in a week’s time? And what about one year from now? The first question is probably very easy to answer, and you’ll most likely make a very accurate prediction. But what about the next question – perhaps this is already starting to get more difficult. For the third question, you may be able to imagine what this looks like, but how true would your imaginings be? Imagination and prediction are not the same. When we think about the future it’s relatively easy to visualise what this might look like; we draw on several aspects to help us: Our past experiences: What are you usually doing on a Tuesday afternoon? Our future hopes and plans: What do you hope to be doing then? Day-to-day influences: What information have we been exposed to? Our sub-conscious: Which lends biases, fallacies, and creative licence. Indeed, we can picture multiple future scenarios and alternatives, which look and feel very plausible. They are based on our experiences and knowledge (plus a dose of our own creativity) so appear to provide a good foundation for predicting the future. And yet, human beings consistently struggle to make good future predictions, even those who have detailed knowledge and experience in their field, such as academics, stockbrokers, and yes, healthcare professionals too. So how do we make predictions with any degree of certainty? Three is the magic number. Our approach to forecasting is based on three key tenets: Predictions are more accurate when considered in a holistic context. Responding to potential future scenarios is easier than building the future. Any prediction has a window of tolerance, so modelling outcomes is key. When it comes to forecasting future drug performance and creating a demand model, we combine a range of approaches to get closer to an accurate prediction. In practice this includes a blend of questioning techniques founded in behavioural science and data science that allows us to refine and model results. These core elements come together to create a powerful demand model solution for our clients: we call it HRW Predict. Predict includes the following steps: Step 1: Ask HRW’s Early Share Estimation Technique (ESET ™) is a proprietary technique whereby future product choices are presented in a congruent choice environment. This provides more realistic responses as respondents need to consider any new product entrants in the context of choosing between existing, familiar and favoured product options. Crucially, this approach addresses the high level of overclaim traditionally seen in demand research resulting from the way new products are presented and which results in two psychological phenomena known as the pedestal effect (an artificial focus on the new product) and unchallenged acceptance (the tendency to focus on positive product attributes without considering the drawbacks). ESET ™ presents the future as if were in the ‘here and now’. Step 2: Refine Once individual uptake figures have been established, the data is recalibrated to refine prescribing estimates, taking into account individual caseloads (and existing sales data when available), as well as developing uptake curves. Step 3: Model Finally, we can model the data. Most commonly, results would be modelled by applying further recalibration metrics taken from explicit questioning in the wider survey. This allows us to provide realistic ranges of potential to assess more and less favourable outcomes (highlighting the degree of risk). In addition, for products in the earlier stages of development, your product profile may not be fully set, so combining the above approaches with conjoint delivers an understanding of how demand might change according to different levels of success across a range of clinical parameters. Bringing all of this together, it is clear that predicting the future demand for a product is not easy, but by utilising a range of techniques, we are confident that we can provide an accurate read. By understanding how people think, behave, and respond, and by using data techniques to refine results, we’re confident that we can provide a far more accurate and considered view of product potential. Apply Now!