A comparison between Fuzzy Linguistic RFM Model and traditional RFM model applied to Campaign Management. Case study of retail business.
MetadataShow full item record
AuthorGonzález Martínez, Rocío; Carrasco, Ramón Alberto; García-Madariaga, Jesús; Porcel Gallego, Carlos Gustavo; Herrera Viedma, Enrique
Martínez, R. G., Carrasco, R. A., García-Madariaga, J., Gallego, C. P., & Herrera-Viedma, E. (2019). A comparison between Fuzzy Linguistic RFM Model and traditional RFM model applied to Campaign Management. Case study of retail business. Procedia Computer Science, 162, 281-289.
Recency Frequency Monetary Value (RFM) is a clear and descriptive way to classify a customer database based on purchasing behavior that direct marketers have used with success since almost the last twenty years. Despite the fashion that exists lately around predictive models and artificial intelligence, direct marketing’s RFM, still has a place in modern database marketing. In a real business environment, RFM can still be useful when models are not practical because it is user friendly and the outcome is always interpretable. It can also be used to combine with other models. In this paper, we show a real example about how easy, accurate and explainable can be a customer segmentation based on the traditional RFM model and the 2-tuple RFM model applied to a customer database. It will help us to better understand the benefits of applying the 2-tuple model instead of the traditional one. We will be able to see how, by applying a k-means clustering on top of the 2-tuple model, segments have a great applicability from the business point of view. By using descriptive variables, we will clarify the cluster description and the model will provide us an extremely clear idea about how customers behave. The main goal for developing this example was to define the best target for a direct campaign communication. Data used for this analysis belongs to a worldwide home furniture, Scandinavian Retailer, and are related to its loyalty program which give us the members’ historical purchase information.