Optimal Allocation of Competitive Marketing Efforts: An Empirical Study

Abstract

There is probably no more difficult and complex problem in marketing than the allocation of the total selling efforts of a firm among the various marketing instruments that influence sales, such as advertising, price, product quality, and personal selling. From a theoretical viewpoint, this problem has been neatly solved and the formal conditions defining an equilibrium have been stated in a number of ways.' Dorfman and Steiner, for example, have derived a general rule which is claimed to be a useful guide for marketing programming; this theorem, however, so far has received little empirical verification.2 Furthermore, implicit in the DorfmanSteiner formula is the assumption that there is no reaction of competitors to the change in marketing inputs made by the firm, while in real market situations mutual interdependence is the rule, since the sales of any firm are also a function of marketing inputs of its rivals.3 The neglect of competitive interdependence is of course a serious shortcoming. The purpose of this article is to show how the Dorfman-Steiner theorem can be expressed in terms of market shares, relative price, advertising, and quality, and how it can then be converted into operational concepts by means of regression analysis, so that attention can be turned to the study of competitors' reactions to change in marketing pressure and especially in advertising. These objectives will be pursued by studying the case of a household durable product sold on three markets of Western Europe characterized by an oligopolistic structure. This paper is divided into three parts. In the first one, a theoretical extension of the Dorfman-Steiner theorem that explicitly considers competitive effects is presented. In the second part, a description of the case studied, of the model, of the data utilized, and of the regression results is given. Using the optimization * Associate professor of marketing research, Louvain University, Belgium. Part of this paper has been presented at the 1968 fall conference of the American Marketing Association. I am grateful to Miss Leboutte for programming assistance, and to K. S. Palda for his advice and comments on an earlier draft. Responsibility for errors remains mine.