Deployment of churn prediction model in financial services industry

Abstract

Nowadays, data analytics techniques are playing an increasingly crucial role in financial services due to the huge benefits they bring. To ensure a successful implementation of an analytics project, various factors and procedures need to be considered besides technical issues. This paper introduces some practical lessons from our deployment of a data analytics project in a leading wealth management company in Australia. Specifically, the process of building a customer churn prediction model is described. Besides common steps of data analysis, how to deal with other practical issues like data privacy and change management that are encountered by many financial companies are also introduced.