Estimando el Customer Lifetime Value: Una Métrica Alternativa para la Segmentación de Clientes aplicado al Sector Retail

Abstract

The success of a company depends on its ability to build and maintain long-term relationships with its most valuable buyers. Ecuadorian companies face the challenge of identifying, differentiating and segmenting their clients. In response to this need, this study seeks to identify an unconventional form of segmentation based on the expected monetary value of customer billing in the retail sector. To this end, a probabilistic model was implemented that is derived from the Pareto/NBD methodology with a Bayesian hierarchy approach to estimate the customer value (CLV), based on a RFM data structure that considers the recency, frequency and monetary value of customers. After the estimate, the respective segmentation of customers was made where 24% of them were found to generate 68% of sales, therefore, three-quarters of customers are considered unprofitable. This research proposes a more efficient segmentation alternative based on customer value in the context of a retail company.