After pointing out some differences between theory and practice in segmenting industrial market, this article reviews the literature on the subject. The latter reveals that too mechanistic or too quantitative segmentation methods are quite inappropriate for the industrial environment where the markets are concentrated and the data sometimes rare. Based on this observation, the article, finally, looks at the problem from another angle by mixing formally a little intuition with the analysis. The end of the article is illustrated by an application of the method. (C) 2000 Elsevier Science Inc. All rights reserved.