Predictive Analytics - The Bottom Line

Overview

In (Walker & Khoshafian, 2010) the case was made for the use of predictive analytics and leveraging them through Business Process Management (BPM) solutions. This companion paper will show the contribution that predictive analytics can make to the bottom line as well as the capabilities of Pega’s products in this space.

Although predictive models are successfully used in many business areas, their value is particularly easy to demonstrate in CRM-related areas. Per Harvard Business School, customer lifetime value (CLTV) takes the following factors into account:

• The base profit (e.g. a subscription rate or annual fee);
• Profit from cross- and up-selling;
• Profit from a price premium (that the customer will pay based on increased confidence in the brand);
• Profit from reduced operating costs (because the customer gets more experienced over time and requires less and less involved service);
• Profit from referrals.

Clearly, the customer experience and related satisfaction play a big role in this formula. A relatively small change in the retention rate has a huge impact on CLTV. In an influential article by (Reichfeld & Sasser, 1990) it was argued that a ‘mere’ 5% improvement in loyalty will increase CLTV by 85% for a bank and 75% for a cards supplier. The other main factor, the success rate of cross-selling new products to existing customers, is similarly linked with customer experience and satisfaction.

In this paper we will consider the business contribution of predictive analytics and focus on improvements in customer insight. We’ll start with predicting customer risk as it does not explicitly feature in Figure 1, yet in some industries, as was made painfully clear over the last few years, it could negate any and all improvements in sales and retention. Following that, case studies are included on sales (cross-selling) and retention.