Knowledge that Transforms

Commentary to Infosino's paper (Infosino, W. J. 1986. Forecasting New Product Sales from Likelihood of Purchase Ratings. Marketing Sci. 5 372–384.).

Author's reply to comments (Morrison, D. G. 1986. Commentary. Marketing Sci. 5 385–386 and Pratt, R. W., Jr. 1986. Commentary. Marketing Sci. 5 387–388.) about his paper (Infosino, W. J. 1986. Forecasting New Product Sales from Likelihood of Purchase Ratings. Marketing Sci. 5 372–384.).

Commentary to Infosino's paper (Infosino, W. J. 1986. Forecasting New Product Sales from Likelihood of Purchase Ratings. Marketing Sci. 5 372–384.).

In a well-known article, Blattberg and Jeuland (Blattberg, R. C., A. P. Jeuland. 1981. A micromodeling approach to investigate the advertising-sales relationship. Management Sci. 27 (September) 988–1005.) apply nonlinear estimation to a market level model based on aggregation of an individual level model of consumer response to advertising. In this paper, a simpler estimation technique is developed and successfully employed for that same model.

Conjoint analysis has achieved widespread acceptance and use among applied market researchers as a tool for forecasting and diagnosing the market performance of new and existing products. Given our reliance on the procedure, those of us who use it are highly receptive to suggestions on how to increase its effectiveness. Hagerty's (Hagerty, M. R. 1986. The cost of simplifying preference models. Marketing Sci. 5 298–319.) paper offers us some provocative new ideas which should force any user of conjoint analysis to reflect upon the models he or she is using and to determine whether there is sufficient evidence to dictate a change.

I am flattered to see that the data from our 1959 experiment are still being analyzed. And I am impressed with the mathematical elegance of Mahajan and Muller's (Mahajan, V., E. Muller. 1986. Advertising pulsing policies for generating awareness for new products. Marketing Sci. 5 89–106.) analysis. It would be presumptuous of me to comment on the technical aspects of the paper. However, some comments on our more recent work in this area may be of interest.

Commentary to Hagerty, M. R. 1986. The cost of simplifying preference models. Marketing Sci. 5 298–319.

Mahajan and Muller (Mahajan, V., E. Muller. 1986. Advertising pulsing policies for generating awareness for new products. Marketing Sci. 5 89–106.) have written a valuable paper that analyzes a particular class of advertising policies under the assumption of an S-shaped response function. As we shall discuss, their results highlight the need for new empirical research to answer crucial questions about what constitutes a pulse and the extent to which S-shaped responses exist.

Awareness forecasting models, such as those discussed in Mahajan, Muller, and Sharma (Mahajan, V., E. Muller, S. Sharma. 1984. An empirical comparison of awareness forecasting models of new product introduction. Marketing Sci. 3 (Summer) 179–197.), will be incomplete until they take account of the awareness effects of mere distribution. Distribution acts as both a main effect and as an interactive partner with advertising (and promotion) in the generation of awareness. Advertising in turn leads to distribution (Heeler et al. [Heeler, R. M., M. J. Kearney, B. J. Mehaffey. 1973. Modelling supermarket product selection. J. Marketing Res. 10 (February) 34–37.]).

Elimination By Aspects (EBA) is a feature-based, psychological processing model of choice whose potential for customer decision modeling has not been exploited. One of several barriers to econometric application of the theory is the lack of an explicit framework for incorporating quantitative variables, such as price. The present study discusses a theoretical treatment of price within EBA which also serves as a guide to the treatment of other quantitative variables. Specifically, it is proposed that prices be represented as a sequence of nested price feature sets, in which the price feature set of an alternative is included in the price feature sets of all lower priced alternatives. The formal consequences of this representation are examined. Some predictions from the theory are tested on customer choice data.