To evaluate the comparative effectiveness of perceptions‐based market segmentation strategies: to what extent do consumers' choice rules and the distinctness and variability of consumer preferences determine the success or failure of PBMS strategies?
The computer simulation is run on an artificial consumer market. Its firm and consumer agents enjoy a certain extent of autonomy and a limited capability of learning. Strategies for incorporating the choice information into the firms' segmentation schemes, consumers' brand choice rules, initial preference patterns and their variability over time are factors in the experimental design.
The market factors “brand choice rule” and “distinctness” and “adaptivity” of preferences significantly influence the profit performance of the segmentation and positioning strategies. The distinctness of the initial pattern of consumer preferences turns out to be least influential while the choice rule is most important.
Computer simulation cannot replace analyses of real‐world data. When, however, advanced explanatory models are made to fit to empirical data the results sometimes are disappointing (and then do not get published). Computer simulation on artificial markets assists in exploring the reasons for success or failure.
Boundedly rational consumers; product classes which are technologically homogeneous and subject to communications‐driven differentiation; consumer preferences that are directly inaccessible and must be inferred from actual brand choice; consumers' perceptions and preferences evolving over time are realistic settings.
Controlling for conditions such as the consumers' choice rules and the distribution and variability of preferences in real markets demands a prohibitive research effort. No empirical study so far has tried to systematically relate the profit performance of marketing strategies to choice rules and preference distinctness and variability.
Mazanec, J. (2006), "Evaluating perceptions‐based marketing strategies: An agent‐based model and simulation experiment", Journal of Modelling in Management, Vol. 1 No. 1, pp. 52-74. https://doi.org/10.1108/17465660610667801Download as .RIS
Emerald Group Publishing Limited
Copyright © 2006, Emerald Group Publishing Limited