!URL http://hbr.org/2007/12/the-flaw-in-customer-lifetime-value/ar/1 !Description Have you ever watched as one of your competitors raked in profits from customers that you had decided not to bother acquiring? Perhaps this happened because your company based its decision on the traditional method of calculating customer lifetime value (CLV). That could be a costly mistake. The standard CLV approach calculates the net present value (NPV) of all anticipated cash flows coming in (revenues) and going out (marketing dollars spent) over some time span (months, years, or even decades) for a given customer. Keep customers who show a positive NPV for the marketing investment, the rules say, and drop the ones who don’t. Those are rational choices at the time of the calculation—but only then. The conventional calculation is thus rather static and actually flawed, because you are not able to factor in a company’s flexibility to cut a given customer loose at any time. Flexibility means options, and options have value. In analyzing the seller’s option to abandon unprofitable customers, my colleagues and I looked at 12 years of purchasing data for more than 100,000 customers of a specialty catalog company. What we found was astonishing: The difference in CLV using the real-options approach versus the traditional approach was, in some cases, as much as 20%. More important, some negative CLVs actually turned positive when the value of the real options was considered. In other words, we found that neglecting a customer with a negative NPV—as traditionally calculated—may be exactly the wrong choice: He could turn out to be highly profitable. Companies have been using real options for a long time to optimize their investment portfolios. It’s time they applied them in the valuation and management of their investment in customers, too. Here’s a five-step process for bringing real-options analysis into the NPV calculation: # Estimate the future purchasing behavior (that is, the probability of purchase and dollar amount expected to be spent) for a set of customers, using the common RFM (recency, frequency, and monetary value) approach. This determines expected revenues. # Calculate costs generated per customer per period. # Use those two inputs to estimate the profit contribution for each customer over the time horizon under consideration, for example ten periods. # For each period, determine whether the expected future profit contribution for that customer might be negative. # Finally, calculate the CLV that includes the option value (the value of abandoning that customer). What would be saved by dropping him at any given period? To give a highly simplified example of this fifth step, imagine a customer that cost \$100 to acquire and \$100 to retain for each subsequent period, thus costing \$500 to keep for five periods. If, based on data from previous customers with similar characteristics, that customer might be expected to purchase \$150 of goods in the first period, \$100 in the next, \$50 the next, and \$0 in the final two periods, a conventional CLV calculation would show that the customer would be unprofitable (he generates just \$300 of revenue at a marketing cost of \$500). But, add in the value of the option of dropping him after the second period (\$100 × 3 = \$300) and the same customer suddenly looks profitable: Over the five periods, the marketing cost is \$200 (\$500 minus the \$300 option to drop) and revenues are \$250 (\$150 for period 1 and \$100 for period 2). In an actual business scenario, this calculation can be mathematically demanding and generally can’t be carried out easily using simple spreadsheet methods. More advanced option-pricing approaches are needed. (For details, see Michael Haenlein, Andreas M. Kaplan, and Detlef Schoder’s “Valuing the Real Option of Abandoning Unprofitable Customers When Calculating Customer Lifetime Value” in the Journal of Marketing, July 2006.) There’s a saying in business: Anyone can save 50% on marketing expenses, but no one knows which 50%. Including real options in the CLV calculation can actually help direct marketing dollars to the right customers.