An empirical comparison of new product trial forecasting models

Abstract

While numerous researchers have proposed different models to forecast trial sales for new products, there is little systematic understanding about which of these models works best, and under what circumstances these findings change. In this paper, we provide a comprehensive investigation of eight leading published models and three different parameter estimation methods. Across 19 different datasets encompassing a variety of consumer packaged goods, we observe several systematic patterns that link differences in model specification and estimation to forecasting accuracy. Major findings include the following observations: (1) when dealing with consumer packaged goods, simple models that allow for relatively limited flexibility (e.g. no S‐shaped curves) in the calibration period provide significantly better forecasts than more complex specifications; (2) models that explicitly accommodate heterogeneity in purchasing rates across consumers tend to offer better forecasts than those that do not; and (3) maximum likelihood estimation appears to offer more accurate and stable forecasts than non‐linear least squares. We elaborate on these and other findings, and offer suggested directions for future research in this area. © 1998 John Wiley & Sons, Ltd.

Building Brands

“…First, we specify word-of-mouth effects to be negligible ( ≈ 0). This specification is consistent with the findings of Hardie et al (1998), who find no word-of-mouth effects across 19 different CPG data sets. Given (1) high variability in weekly sales arising from weekly promotions, (2) the fact that most products where t indicates the base sales for the brand at time t and is the base market potential.…”

How to project customer retention

“…However, we feel that such a conclusion is premature. After all, in other areas of marketing there are plenty of models that have been used to provide accurate forecasts of the behavior of a cohort of customers beyond the range of observations (e.g., Hardie, Fader, & Wisniewski, 1998, for the case of new-product-sales forecasting). Thus, in the next section, we formulate a probabilistic model of contract duration that is based on a simple "story" of customer behavior.…”

“…Rather, we should use its continuous-time analog, the exponential-gamma (EG) distribution (also known as the Lomax distribution or the "Pareto distribution of the second kind"). Such a model assumes that the duration of an individual customer's relationship with the firm is characterized by the exponential distribution, and that heterogeneity in "departure rates" is captured by a gamma distribution (Hardie et al, 1998;Morrison & Schmittlein, 1980). Models for noncontractual settings are more complicated because the time at which a customer becomes inactive, and the likelihood that it has occurred at all, must be inferred from the transaction history.…”

Modeling of a Small Transportation Company’s Start-Up with Limited Data during Economic Recession

“…Most modeling and forecasting approaches with limited data are about rapidly changing industries like motion pictures, telecommunications, or new products with a short history [9][10][11][12][13]. They suggest that combining ARIMA and diffusion models can improve one-year-ahead predictions, especially in the high technology market.…”