This commentary reports some of the limitations of Ehrenberg's approach, in particular with respect to the lack of reproducible parameter estimates and the inadequacy of the Dirichlet model to capture the "double jeopardy" effect. It explains this failure by the varying distribution levels across brands, inconsistent with the assumption embedded in the standard Dirichlet model.
BEMMAOR, A.C. (1994). Commentary on ASC Ehrenberg's "Theory or Well-Based Results: Which Comes First?". In: G. LAURENT, G.L. LILIEN, B. PRAS eds. Research Traditions in Marketing. 1st ed. Boston, MA: Kluwer Academic Publishers, pp. 109-115.