milad-deriv/customer_lifetime_value

Predicting Customer Lifetime Value (CLV) in non-contractual setting.

Customer Lifetime Value (CLV)

Predicting customer lifetime value in non-contractual situations in which customers can make purchases at any time.

Predicting customer lifetime value and future purchases

In most cases value of a firm is profits from existing and future customers (a.k.a. Customer Equity). Research done by Frederick Reichheld of Bain & Company (the inventor of the net promoter score) shows increasing customer retention rates by 5% increases profits by 25% to 95% (Reichheld 2001).

It is possible to calculate Customer Equity (CE) because Customer Lifetime Value (CLV) can be measured with a reasonable degree of precision.

CLV is the present value of the future (net) cash flows associated with the customer (Gupta and Lehmann 2003). It is a forward-looking concept, not to be confused with historic customer profitability.

Not all customers are equally important to a firm. Maintaining long-term relation with all of them (especially the loss makers) is not optimal because eventually marketing is all about attracting and retaining profitable customers (Kotler and Armstrong 1996). Hence the objective of CLV is firstly on general topics of firm’s profitability and secondly as an input in customer acquisition decision and customer acquisition/retention trade-offs (Berger and Nasr 1998).

In every business it is important to find the most valuable customers. The Pareto Principle says 20% of your customers represent 80% of your sales. So it would be a great opportunity for the company, if we could identify that 20%, not only historically, but in the future as well. CLV can help us to identify these customers.

Objectives

The primary goal of this work is to build a probabilistic model for forecasting customer lifetime value in non-contractual setting on an individual level.

Using the results of this exercise, managers should be able to:

• Distinguish active customers from inactive customers.
• Generate transaction forecasts for individual customers.
• Predict the purchase volume of the entire customer base.