Knowledge that Transforms

The purpose of this note is to supplement some recent articles which discuss correlations between two grouped random variables. In this paper we give a more parsimonious and less computationally complex method of assessing the effect of grouping while also giving some insights into the best ways to group continuous variables. Implications of these results for certain marketing studies are given.

A decision support system for planning marketing strategies and allocating resources for a multi-store retailer is described. This decision support system combines well-known model building and analysis methodology, sophisticated computer software, and attention to management's implementation needs in order to apply management science thinking to messy, high level strategy and forecasting problems. The system consists of a planning model, national campaign evaluation system, experimental analysis system, and an ongoing interactive data base and reporting system. The planning model was first implemented with subjective judgment as input. The rest of the system was then refined to aid management in improving their subjective judgments, and for tracking and control. The marketing mix for total and business entity (groups of departments) sales is presently being planned with the support of this system. Profit improvements have been identified and implemented.

This paper describes and evaluates the application of a dynamic stochastic model of consumer response. The model describes, then forecasts, how consumers respond to a new transportation service and to the marketing strategies used during its introduction. The model is estimated on survey data during the first 11 weeks of service. Forecasts over the next 19 weeks are then compared to actual ridership as measured by dispatch records. The model is simple. At any point in time, consumers are described by a set of “behavioral states”, indicating (1) whether they are aware of the new service (DART) and (2) what mode of transportation was used for their last trip. Behavior is described by movement among behavioral states. E.G., If a car user tries DART, he makes a transition from ‘car used for last trip’ to ‘DART used for last trip’. The transition probabilities and the rate of transition are dependent on marketing strategies (direct mail, publicity), word of mouth, consumer perceptions, availability of a mode, and budget allocation to transportation. The advantages and disadvantages of the model and the measurements are discussed with respect to predictive ability and managerial utility.

Longitudinal changes in perceptions of both new and existing brands in the same product class are studied. Specifically pairwise similarity judgments are collected before and after participation in a 6-occasion choice and usage experiment. Comparisons are made across both the original similarity jugments and the resulting group-level perceptual spaces. Changes are a function of type of brand (old or new) and experimental manipulations. Substantively, the results support the previously untested configural invariance hypothesis, i.e., the perceptions of existing brands are not substantially changed by the introduction of new brands that are relatively different. This suggests this type of scaling technique can be used to predict consumer reactions to new product introductions—even if they are fairly different from current offerings.

A two state Markov chain model is used to describe and forecast the over time behavior of the best retail customers at Merrill Lynch. This model has 4 behaviorally meaningful parameters which capture the effect of recently being a prime customer, the differing average commissions generated across customers and the exiting of some of these customers from the Merrill Lynch system. This model helps management to understand the dynamics of the prime customers' behavior. In particular, the forecasts generated by the model allow for better analyses of possible strategies for providing special services for these very good customers. The model which was developed with 1976–1979 data is validated against the actual 1980 behavior of Merrill Lynch customers.

The experience curve phenomenon of falling marginal costs associated with accumulated output or production experience has given rise to dynamic pricing models. Optimal pricing policies will depend upon the nature of the dynamic demand and cost functions. In this note we shall show that the demand function employed by Bass (Bass, F. M. 1980. The relationship between diffusion rates, experience curves, and demand elasticities for consumer durable technological innovations. J. Bus. 53 (July) S51–S67.) when taken in conjunction with the experience curve cost function leads to a multiperiod pricing strategy which is always less than the myopically optimal price. Further, we present a dynamic programming algorithm for the multiperiod strategy in which we have explored the effects on discounted profits of myopic pricing versus multiperiod pricing. The results of this comparison may, to some, be somewhat surprising and may have managerial significance.

This paper provides a mathematical framework for modeling demand and determining optimal price schedules in markets which have demand externalities and can sustain nonlinear pricing. These fundamental economic concepts appear in the marketplace in the form of mutual buyers' benefits and quantity discounts. The theory addressing these aspects is relevant to a wide variety of goods and services. Examples include tariffs for electronic communications services, pricing of franchises, and royalty fees for copyrighted material and patents. This paper builds on several previous results from microeconomics and extends nonlinear pricing to markets with demand externalities. The implications of this price structure are compared to results obtained for flat rates and two part tariffs in a similar context. A case study is described in which the results were applied to planning the startup of a new electronic communications service.

Structural transformations of the MCI model are presented which make the model easily estimated using dummy variables with widely available regression packages. The MCI model is empirically shown to provide better predictive power than several other models of similar form, but ones which do not produce logically consistent market share estimates.

Modeling efforts in the area of new product introductions have had a significant impact on marketing planning and strategy. One result of these efforts, BBDO's New Product Early Warning System (NEWS), has been used since the late 1960s to provide marketing managers with forecasts and diagnostic information regarding their new product strategies. This article presents the specification of the NEWS model, its parameter estimation methods, and its validation. A brief case history is also included which illustrates how the model is applied in a typical new product situation. NEWS is designed to use a variety of readily obtainable input data to generate forecasts of consumer awareness, trial, repeat purchase, usage, sales, and market share for a new brand. These outputs, combined with diagnostics from the model, can then be incorporated into the marketing plan in a way that will improve the new entry's chances of success in the marketplace. The model can be used to project early test market data (NEWS/Market); or it can be used to analyze pre-test market data (NEWS/Planner).