Note---A Comment on "Modeling the Marketing Mix Decision for Industrial Products"

A recent article by Lilien (Lilien, Gary L. 1979. ADVISOR 2: Modeling the marketing mix decision for industrial products. Management Sci. 25 (February) 191--204.) reports the principal findings of the ADVISOR 2 project, in which regression models are used to explain levels of advertising and marketing expenditures by industrial product manufacturers. Lilien compares ADVISOR results with some analyses of PIMS data which we reported in 1976 (Buzzell, Robert D., Paul W. Farris. 1977. Marketing costs in consumer goods industries. Hans Thorelli, ed. Strategy + Structure = Performance. Indiana University Press, 122--145.), and states that "... the results of the Buzzell-Farris study on the PIMS data and the results of ADVISOR are quite consistent." We cannot agree with this conclusion. We believe there are important inconsistencies between the two studies, which should be considered by anyone who might use the ADVISOR model---or the PIMS-based models---to evaluate actual marketing budgets.