In this paper we consider the use of correspondence analysis (CA) of rating data. CA of rating data allows a joint representation of the rated items (e.g. attributes or products) and individuals. However, as the number of individuals increases, the interpretation of the CA map becomes difficult. To overcome this problem, we propose a method that allows the depiction of additional variables, for example, background characteristics that may be of interest in identifying consumer segments, in the CA map. The idea we use is based on the representation of supplementary variables in ordinary CA. However, as the format of the additional variables is typically different from the rating data, a recoding is required. We illustrate our new method by means of an application to data of a product perception study for five cream soups.