Aplicación de un modelo de segmentación de cartera pendiente para la personalización de estrategias de cobranza. Caso de estudio: industria dedicada a la comercialización de cemento de la ciudad de Cuenca, Ecuador

Abstract

This research aims to apply a pending portfolio segmentation model based on the characteristics and variables derived from each client’s commercial transactions over a defined period. Drawing on related studies and the author's experience in the sector, the analysis is based on data from a company dedicated to the production and commercialization of cement in the city of Cuenca, Ecuador. The project is justified by proposing solutions grounded in the collection, processing, scaling, and standardization of information related to various economic movements and interactions. This information, combined with the use of the K-means algorithm, defines segments that, due to their unique characteristics, allow the assignment of collection strategies tailored to each group’s specific needs. As a result, six standout groups or clusters were identified, which should be considered for the implementation of well-defined collection techniques and targeted resource allocation. As a final contribution, specific communication and collection strategies were proposed for each group, aiming to improve accuracy in the recovery of company capital. In conclusion, implementing a segmentation model allows for a deeper understanding of customer behavior, changes in their attitudes, and their likelihood of becoming a financial risk; this facilitates strategic decision- making that contributes to reducing delinquency and improving the company's financial performance.