On Improving the Effectiveness of Test Marketing Decisions

While simulation and controlled pretest‐marketing approaches are increasingly used, the traditional test marketing is still the most widely used approach for product and marketing programme testing. For several decades now, the most popular research and professional approach to test market selection has involved the search for an “average” America for test marketing. Discusses the problems associated with this approach and argues that the US is very diverse regionally and, consequently, “middletowns” do not exist. Recommends a multivariate approach for city segmentation to develop homogeneous sets of cities from which test markets may be selected for control and testing purposes. The study also discusses the empirical findings and managerial implications along with ways to validate the recommended approach.