Modelling customer churn for the retail industry in a deep learning based sequential framework

TL;DR

This paper introduces a deep learning-based sequential survival model to predict customer churn in retail, enabling personalized predictions without extensive feature engineering.

Contribution

It presents a novel deep survival framework using recurrent neural networks to model individual customer churn risk based solely on behavioral data.

Findings

Effective prediction of customer churn risk.

Avoids complex feature engineering.

Personalized survival models for retail customers.

Abstract

As retailers around the world increase efforts in developing targeted marketing campaigns for different audiences, predicting accurately which customers are most likely to churn ahead of time is crucial for marketing teams in order to increase business profits. This work presents a deep survival framework to predict which customers are at risk of stopping to purchase with retail companies in non-contractual settings. By leveraging the survival model parameters to be learnt by recurrent neural networks, we are able to obtain individual level survival models for purchasing behaviour based only on individual customer behaviour and avoid time-consuming feature engineering processes usually done when training machine learning models.