Estimating the Confidence Interval for the Optimal Marketing Mix: An Application to Lead Generation

The Dorfman-Steiner Theorem has provided the marketing community with a powerful result for allocating resources between competing marketing mix variables. It is well known that the optimal allocation of resources is in direct proportion to their demand elasticities. To implement this result, the marketing manager must know the elasticities of the various marketing elements under his/her control. Since the precise values of these elasticities is rarely, if ever, known, the manager must use estimates of the elasticities in allocating the resources. Point estimates of the elasticities can be obtained from laboratory or field experiments and from econometric models. In both cases, these estimates are known with uncertainty. This paper discusses the appropriate method for incorporating uncertainties in the point estimates of the elasticities to yield rigorous confidence intervals applicable to the ratio of the elasticities. An empirical example is used to illustrate the methodology.