A stochastic, dynamic model of advertising, which incorporates both advertising and word-of-mouth effects, is formulated. The time between the acquisition of new customers is assumed to be random. The distribution of the time until the firm obtains a new customer depends upon the rate of advertising expenditures and upon a word-of-mouth parameter. The problem of choosing the rate of advertising expenditures so as to maximize long-run expected profit is formulated as a continuous-time Markov decision chain. The impact of changes in various parameters of the model on optimal advertising decisions is studied.
This paper examines the feasibility, practicality, and predictive ability of the consumer model which was proposed by Hauser and Shugan (Hauser, J. R., S. M. Shugan. 1983. Defensive marketing strategies. Marketing Sci. 2 (4, Fall) 319–360). We report results in two product categories, each representing over $100 million in annual sales. We develop “per dollar” perceptual maps and empirical consumer “taste” distributions. As a first test of the model, we compare the predictive ability of the consumer model in one category to (1) pretest market laboratory measurement models, (2) traditional perceptual mapping procedures, (3) a hybrid model using price as an attribute, and (4) actual market shares in test market cities. In the second product category, we illustrate the application of the quantitative model to augment managerial judgment. Besides developing an empirical version of the “Defender” consumer model, our analyses raise a number of behavioral hypotheses worth further investigation.
This paper extends previous single equation econometric modeling efforts of the “sales-advertising” relationship in elections. Operating within the perspective of the general demand context of voting behavior, it develops a comprehensive electoral market model that takes account of potential simultaneous relationships among votes cast, campaign expenditures, and voting participation (turnout). The model is estimated using 3SLS techniques with aggregate electoral district data from each of eight Canadian provincial elections. The model appears to yield consistent and therefore generalizable empirical results across the eight election events. Chief among the empirical results is the expected relationship between campaign efforts and voting behavior: “own” communications efforts (as proxied by campaign expenditures) have a positive impact on votes received, while competitive communications efforts draw votes away from a candidate.
This article presents a new approach to target marketing. An industrial market for electrical equipment is segmented on the basis of the strength of current preferences as defined by the probabilities of selecting the various suppliers. A modification of a disaggregate attribute choice model (logit) is used to identify the firm's “switchable” customers. The “switchable” customers are then targeted by direct personalized mail and visits by a “missionary” sales force. A major supplier of large-scale electrical equipment utilized this methodology with impressive sales results.
Five awareness forecasting models embedded in their respective new product introduction models are compared. Conditions which govern the differences in the awareness estimates provided by the various models are delineated. Managerial implications of the results are discussed.
A model of information search is proposed which assumes that a consumer chooses the next piece of information so as to maximize his expected value of sample information. The cost of processing, the perceived correlation between attributes, and the perceived importance of attributes would all affect information choice. Three sets of propositions are derived. The model is also estimated and tested for subjects performing an information display board task.
This paper investigates the order of the brand choice process at the individual family level using a variety of tests: t, likelihood ratio, binomial runs, and multinomial runs tests. The test of stationarity developed permits separate analyses for stationary and nonstationary purchase sequences. Our analyses indicate that the purchase sequences of a majority of stationary consumers are consistent with the zero-order assumption. This result is observed for all the nine frequently-bought packaged goods studied with the use of panel data and for a product category (soft drinks) using experimental data.
A model for the prediction and explanation of individual television viewing choice is presented, incorporating considerations of utility, audience flow, and audience segmentation. The proposed model provides a quantifiably explicit theoretical explanation of television viewing choice, and its validation on large-sample network viewing data provides a baseline degree of accuracy against which the performance of future television viewing models may be compared. Of direct relevance to advertising agencies and the television networks is the suitability of the model for estimating the comparative impact of alternative programs on the audience size and composition of competing programs in the immediate and subsequent time slots.
An accurate understanding of the structure of competition is important in the formulation of many marketing strategies. For example, in new product launch, product reformulation, or positioning decisions, the strategist wants to know which of his competitors will be most affected and hence most likely to respond. Many marketing science models have been proposed to identify market structure. In this paper we examine the managerial problem and propose a criterion by which to judge an identified market structure. Basically, our criterion is a quantification of the intuitive managerial criterion that a “submarket” is a useful conceptualization if it identifies which products are most likely to be affected by “our” marketing strategies. We formalize this criterion within the structure of classical hypothesis testing so that a marketing scientist can use statistical statements to evaluate a market structure identified by: (1) behavioral hypotheses, (2) managerial intuition, or (3) market structure identification algorithms. Mathematically, our criterion is based on probabilities of switching to products in the situation where an individual's most preferred product is not available. ‘Submarkets’ are said to exist when consumers are statistically more likely to buy again in that ‘submarket’ than would be predicted based on an aggregate “constant ratio” model. For example, product attributes (e.g., brand, form, size), use situations (e.g., coffee in the morning versus coffee at dinner), and user characteristics (e.g., heavy versus light users) are specified as hypotheses for testing alternate competitive structures. Measurement and estimation procedures are described and a convergent approach is illustrated. An application of the methodology to the coffee market is presented and managerial implications of six other applications are described briefly.
In this paper our previous work on monopoly and oligopoly new product models is extended by the addition of pricing as well as advertising control variables. These models contain Bass's demand growth model, and the Vidale-Wolfe and Ozga advertising models, as well as the production learning curve model and an exponential demand function. The problem of characterizing an optimal pricing and advertising policy over time is an important question in the field of marketing as well as in the areas of business policy and competitive economics. These questions are particularly important during the introductory period of a new product, when the effects of the learning curve phenomenon and market saturation are most pronounced. We consider first the monopoly case with linear advertising cost, exponential demand, and three different pricing rules: the optimal variable pricing, the instantaneous marginal pricing, and the optimal constant pricing rules. Several theoretical results are established for these rules including the facts that the instantaneous marginal pricing rule is a myopic version of the optimal pricing rule and the optimal constant pricing rule is a weighted average over time of the instantaneous marginal pricing rule. Another surprising result is that, after the market is at least half saturated, a pulse of advertising must be preceded by a significant drop in price. Numerical solutions of a number of examples are discussed. Oligopolistic models are analyzed as nonzero-sum differential games in the rest of the paper. The state and adjoint equations are easy to write down, but impossible to solve in closed form. Hence we describe how to reformulate these models as discrete differential games, and give a numerical algorithm for finding open loop Nash solutions. The latter was used to solve three triopoly models. In each case it was found that optimal prices and advertising rates start high and steadily decline.