Business Performance Management System for CRM and Sales Execution 1 Markus Ettl, Bianca Zadrozny, Pawan Chowdhary and Naoki Abe IBM T.J. Watson Research Center, Yorktown Heights, NY 10598, USA {msettl,zadrozny,chowdhar,nabe}@us.ibm.com 1 Submitted to: International Workshop on Business Process Monitoring & Performance Management (BPMPM 2005). Abstract In 2004 the IBM Telesales organization launched a new customer segmentation process to improve profits, revenue growth and customer satisfaction. The challenges were to automatically monitor customer segment status to ensure results are in line with segment targets, and to automatically generate high- quality predictive analytical models to improve customer segmentation rules and management over time. This paper describes a software solution that combines business performance management with data mining techniques to provide a powerful combination of performance monitoring and proactive customer management in support of the new telesales business processes. 1. Introduction As companies adopt more customer-centric business processes, customer segmentation strategies are the key to balancing the delivery of optimal customer experiences with the need to maximize long- term customer revenue and loyalty. However, developing an effective customer segmentation strategy is a serious challenge for many companies since they often lack the ability to track and interpret the huge volumes of customer data that exist throughout the enterprise. Many companies also lack real insight into their customer base to detect changes in overall buying behavior and customer value which makes it impossible to evaluate the long-term success of their customer segmentation efforts. Business Performance Management (BPM) enables a company to keep up with a rapidly changing business environment. In its quest to preserve or expand competitive advantages that are threatened by intense competition and changing customer preferences, a company needs to continuously monitor critical business processes, detect existing or anticipated business issues, and respond swiftly to opportunities and threats [5,8,9]. To do so, advanced business process monitoring capabilities and decision support are essential. In this paper we describe a BPM system developed for the IBM Telesales organization to support Customer Relationship Management (CRM) and sales execution. Our BPM solution provides an analytical and information infrastructure to support the segment strategy. It automates the demanding task of monitoring customers and telesales center performance with predefined business metrics and customer segments. Customer segments are defined by customer behavior, demographic attributes and customer value. Business metrics include profitability and sales targets, recency and frequency of purchases, and monetary value scores. Our analytical modeling techniques help telesales executives create more accurate customer segments and predict which customers are likely to churn which in turn improves profitability, customer satisfaction and revenue contributions. The system leverages IBM’s BPM architecture [2,3] to develop an end-to-end performance management framework. It comprises several components. First, a data warehouse that stores several years of historical customer purchases (laptops, desktops and options) on several thousand customer accounts and millions of sales transactions; second, an ETL (data extraction, transformation and loading) module that provides daily access to a production data mart to load new sales transactions into the data warehouse; third, a web- based management dashboard that captures key operational metrics indicating the effectiveness of the segmentation model against planned business objectives; fourth, a novel application of predictive modeling technology that is optimized for discovering recommendations for customer advancements through the segments based on previous customer spend and analytical correlations. These models help realize an