The objective of this article is to present a new statistical approach for the study of consumer liking. Total liking data are extended by incorporating liking for specific sensory properties. The approach combines different analyses for the purpose of investigating the most important aspects of liking and indicating which products are similarly or differently perceived by which consumers. A method based on the differences between total liking and the specific liking variables is proposed for studying both relative differences among products and individual consumer differences. Segmentation is also tested out in order to distinguish consumers with the strongest differences in their liking values. The approach is illustrated by a case study, based on cheese data. In the consumer test consumers were asked to evaluate their total liking, the liking for texture and the liking for odour/taste. (C) 2012 Elsevier Ltd. All rights reserved.