Preference Learning and Demand Forecast
Marketing Science, forthcoming
42 Pages Posted: 19 Aug 2019 Last revised: 15 May 2020
Date Written: August 2, 2019
Understanding consumer preferences is important for new product management, but is famously challenging in the absence of actual sales data. Stated-preferences data are relatively cheap but unreliable, whereas revealed-preferences data through actual choices are reliable but expensive to obtain prior to product launch. We develop a cost-effective solution. We argue that people do not automatically know their preferences, but can make an effort to acquire such knowledge when given sufficient incentives. The method we develop regulates people's preference-learning incentives using a single parameter, realization probability, meaning the probability with which an individual has to actually purchase the product she says she is willing to buy. We derive a theoretical relationship between realization probability and elicited preferences. This allows us to forecast demand in real purchase settings using inexpensive choice data with small to moderate realization probabilities. Data from a large-scale field experiment support the theory, and demonstrate the predictive validity and cost-effectiveness of the proposed method.
Keywords: preference elicitation, demand forecasting, incentive alignment, choice experiment, field experiment, external validity
Suggested Citation: Suggested Citation