The authors define a market segment to be a group of consumers homogeneous in terms of the probabilities of choosing the different brands in a product class. Using the matrix of joint probabilities of switching from one brand to another obtained from panel data, segments are extracted using the optimization techniques developed for estimating Latent Class Models. Since a segment can be characterized by the group of brands having “large” choice probabilities, the approach identifies competitive market structure by the possibly overlapping groups of brands corresponding to the segments. An application to the Instant Coffee market is presented.