Data-driven decision making in football clubs

: understanding member churn at FC Porto

Abstract

The football industry has become increasingly reliant on data to understand fan behavior and drive strategic decisions. In this context, member retention is a critical challenge for football clubs, as memberships represent a key source of revenue and long-term engagement. Although churn prediction models are established in sectors such as telecommunications, banking or retail, their application in the sports context remains insufficiently explored. This thesis investigates membership cancellations at Futebol Clube do Porto and develops predictive models to anticipate churn. By analyzing demographic, behavioral, and payment-related variables, the study applies machine learning techniques to identify members with a higher probability of leaving. This project is based on real data from six seasons. Using the CRISP-DM methodology and the application of machine learning models, namely Decision Tree, Random Forest, Logistic Regression and Neural Network, the objective is to identify the main variables that explain member churn. This research contributes to a deeper understanding on churn in sports organizations and supports data-driven decision-making in fan relationship management and offers practical insights that can underpin more effective, data-driven retention strategies.

Date of Award	4 Jul 2025
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Vera Lúcia Miguéis Oliveira e Silva (Supervisor)