How Should You Calculate Customer Lifetime Value?
Should marketers subtract the cost of acquiring a customer before assessing that customer’s lifetime value?
Customer lifetime value (CLV), which is the present value of cash flows from a customer relationship, can help managers make decisions regarding investments in customer relationships.1 For example, a marketer might use CLV to decide whether to spend marketing dollars to acquire new customers or to increase the retention rate of existing customers. CLV can be difficult to calculate because it often relies on the ability to predict future customer retention rates.2 However, we think one major source of confusion among marketers — whether to include customer acquisition cost in the CLV calculation — can be easily avoided. CLV is easier to understand, and in our view more useful, if marketers don’t subtract the acquisition cost from their calculation of CLV before reporting it.3 To be sure, customer acquisition costs are a major item in marketing budgets. Such costs should affect decisions as to whether to pursue prospective customers. But this does not mean that acquisition costs need to be subtracted from CLV before the value of the customer is reported.
CLV is often used to measure the value of customers who have already been acquired. The acquisition costs have therefore already been incurred. Even if the company made a mistake in acquiring a customer and the acquisition costs exceeded the customer’s value, knowledge of this cannot change the earlier acquisition decision. Acquisition costs are “sunk” and should be ignored when making forward-looking decisions.4
Many marketers persist in subtracting acquisition costs before reporting CLV, which results in several ongoing problems. The majority of marketers we surveyed thought that customers with the same value going forward had the same CLV. However, this is not true when acquisition costs are subtracted from CLV before CLV is reported. A highly profitable customer can appear to have the same value as a less profitable customer if the highly profitable customer cost more to acquire.
References
1. R. Venkatesan and V. Kumar, “A Customer Lifetime Value Framework for Customer Selection and Resource Allocation Strategy,” Journal of Marketing 68, no. 4 (October 2004): 106-125.
2. There are many problems with assessing CLV, but we are not able to address all of them here, primarily for reasons of space. In brief, when used as a prediction of the future, CLV measurement is challenging. For example, it can be extremely difficult to project for customers with whom the company does not have a contract or even a regular amount of revenue. It is tough to know whether a customer has been retained but is an irregular purchaser, or whether the customer will never buy again. Discount rates are hard to estimate and, given that they change with risk, should theoretically differ between customers. The problems of calculation can be especially challenging given customer heterogeneity. Further, even if you can calculate CLV, what to do with it can be a challenge, as differentially serving customers can be controversial. For further discussion of these issues, see, respectively, J. Romero, R. van der Lans, and B. Wierenga, “A Partially Hidden Markov Model of Customer Dynamics for CLV Measurement,” Journal of Interactive Marketing 27, no. 3 (August 2013): 185-208; P.S. Fader, B.G.S. Hardie, and K. Jerath, “Estimating CLV Using Aggregated Data: The Tuscan Lifestyles Case Revisited,” Journal of Interactive Marketing 21, no. 3 (2007): 55-71; P.S. Fader and B.G. Hardie, “Customer-Base Valuation in a Contractual Setting: The Perils of Ignoring Heterogeneity,” Marketing Science 29, no. 1 (January-February 2010): 85-93; and C. Homburg, M. Droll, and D. Totzek, “Customer Prioritization: Does It Pay Off, and How Should It Be Implemented?” Journal of Marketing 72, no. 5 (September 2008): 110-130.
3. P.E. Pfeifer, M.E. Haskins, and R.M. Conroy, “Customer Lifetime Value, Customer Profitability, and the Treatment of Acquisition Spending,” Journal of Managerial Issues 17, no. 1 (spring 2005): 11-25.
4. There is an active stream of research on sunk costs and the fact that people inappropriately consider sunk costs in their decisions, thus exhibiting sunk cost bias. One of the most popular demonstrations of sunk cost bias involves coaches in the NBA giving more time than players’ performance warrant to players picked earlier in the draft. The argument is that draft pick “cost,” an early pick, is sunk, yet coaches continue to give these players court time to justify that cost. See B.M. Staw and H. Hoang, “Sunk Costs in the NBA: Why Draft Order Affects Playing Time and Survival in Professional Basketball,” Administrative Science Quarterly 40, no. 3 (September 1995): 474-494.