Forecasting new customers' behaviour by means of a fuzzy unsupervised method

The use of unsupervised fuzzy learning classifications techniques allows defining innovative classifications to be applied on marketing customer's segmentation. Segmenting the clients' portfolio in this way is important for decision-making in marketing because it allows the discovery of hidden profiles which would not be detected with other methods. Different strategies can be established for each defined segment. In this paper a case study is conducted to show the value of unsupervised fuzzy learning methods in marketing segmentation, obtaining fuzzy segmentations via the LAMDA algorithm. The use of an external decision variable related to the loyalty of the current customers will provide useful criteria to forecast potentially valuable new customers. The use of the introduced methodology should provide firms with a significant competitive edge, enabling them to design and adapt their strategies to customers' behaviour.