Being proactive — analytics for predicting customer actions

Customers should be at the heart of most businesses, and in particular service providers such as BT. In order to serve our customers better, we regularly introduce reliable processes and procedures to improve interaction with our customers, which is known as customer relationship management. Typically, organisations collect and keep large volumes of customer data as part of their processes. Analysis of this data by business users often leads to discovery of valuable patterns and trends that otherwise would go unnoticed and that can lead to prioritisation of decisions on future investments. Current tools available to business users are limited to visualisation and reporting of data. What is needed is modelling customer behaviours to be able to build future scenarios. More advanced tools and techniques have been available for a number of years but have not been developed for the business community due to the level of expertise required to use them. In this paper we present a number of tools and techniques developed for business users to perform advanced analysis on customer data. The tools can be used to perform sensitivity analysis, what-if analysis and impact analysis, all of which are aimed at prediction and simulation of future customer actions. The paper also covers application of the tools to real customer data and reports on some of the results obtained.