What to Read Next
Traditional marketing practices are increasingly viewed with skepticism. In many organizations, marketers struggle to document the return on investment for marketing expenditures; as a result, the marketing function is poorly aligned with the strategic goals of the company, marketing has less influence in the boardroom — and the marketing budget allocation is viewed as a questionable cost rather than a worthy investment.
To negate such criticisms, marketers need to realign their role and redefine the scope of marketing so that it can directly relate to strategic outcomes for the company. One way to achieve this goal is through extensive application of predictive analytics, in both the formulation of marketing strategies and customer management. Along these lines, a study we conducted suggests that certain types of marketing efforts — those developed using analytics to identify customers’ lifetime value to a company — can create shareholder value and influence stock prices in a predictable fashion.
We began our research by analyzing more than five years of customer information from two Fortune 1000 companies, one selling to businesses and the other selling directly to consumers. The customer databases of both companies contained rich information related to customer transactions, marketing communications and customer-level characteristics that could potentially influence the customer lifetime value of each customer to the company. However, neither of the two companies had employed the CLV metric, which is defined as the discounted net present value of expected future cash flows from a customer.
V. Kumar and D. Shah, “Expanding the Role of Marketing: From Customer Equity to Market Capitalization,” Journal of Marketing 73, no. 6 (November 2009): 119-136.
As a result, one of our first steps with each of the two companies entailed developing a model to compute the lifetime value of that company’s customers, using as much customer-level information as possible from the company’s customer database. The lifetime value for each customer in this study was calculated as the net present value of the expected cash flows from that customer over the next three years. (Although the time horizon of three years may not reflect the “lifetime” duration of the customer, it captures the majority of the customer’s true lifetime value given the fact that the computation is based on the discounted cash flow approach. As a result, it is a common practice to compute CLV at the individual customer level based on the expected cash flows over the next three years.) We then computed the sum of all customers’ CLV, which represented the net present value of expected future cash flow streams from existing customers, plus some estimates of projected discounted cash flow streams from future customers that the company expected to acquire during that time. (The detailed results of our study, including computation details, were reported in the November 2009 issue of the Journal of Marketing. See “Related Research.”)
We then developed a link function to relate this sum of all the customers’ lifetime value to the market capitalization of the company, as indicated by the stock price. If customers are the fundamental drivers of cash flow for a business, the sum of the CLV metric should relate well with the business’s value as indicated by its stock price. In practice, however, the relationship is weak at best, because the company valuation based on the stock price is risk-adjusted, but the customer valuation typically is not.
In other words, the CLV model estimates the timing and value of future cash flows but does not account for the inherent risks associated with the future stream of cash flows from customers. Cash flow risks can render the future revenue stream vulnerable or volatile, thereby affecting a company’s valuation. For example, if two companies have projected future cash flows of equal value, the company with lower perceived risk (i.e., future cash flows with expected lower volatility and vulnerability) will get a higher market valuation than the organization with higher perceived risk.
The implication for marketing is to account for cash flow risks when estimating the future cash flow streams from customers. The vulnerability of customer cash flow streams can be measured in terms of the propensity of a customer to defect in the future. That can be estimated through measures such as customer satisfaction, customer loyalty, relative switching costs and willingness to repurchase in the future. The volatility of customer cash flow streams can be measured in terms of the relative stability of the cash flow streams. Our research suggests that, by accounting for these risks in the CLV computation, companies can substantially strengthen the relationship between their marketing outcomes and their stock market value.
To test the effectiveness of our approach, we conducted a field experiment over a period of nine months. We designed specific marketing strategies and differentiated them across customers, based on the CLV of each customer or customer segment. With the active support of top management from both of the companies we studied, the marketing strategies were implemented and the performance outcome was measured in terms of the increase in the customers’ CLV. By applying the link function we had developed, a corresponding lift in stock price for both companies was predicted and then compared with actual stock prices over the nine-month duration of the field experiment. We also compared the two companies’ stock market performance over that period with that of their major competitors.
The results were astonishing. While implementing these “next-generation” marketing strategies, the two companies saw their stock prices increase by 33% and 58% — increases in their stock market valuation that were substantially greater than those seen by their three top competitors during that period. What’s more, using the link function relating the CLV of a company’s customers to its stock market valuation, we were able to predict the two companies’ stock price increase within an error range of 12% to 13% over a nine-month period.
Segmenting by CLV
The business-to-business company we studied (which we’ll call “TechInc,” not its real name) is a high-tech manufacturer selling computer-related hardware and supporting software to business establishments. The consumer-oriented company (which we’ll call “FashInc”) is a large national retail chain selling ready-to-wear apparel, shoes and accessories. In both companies, the difference in the value of each customer was determined by quantifying the drivers of customer value that varied across customers, such as customer-specific characteristics, company-initiated marketing interventions and transaction behavior. The CLV computation helped in determining the relative contribution of each customer to the company’s bottom line. In both TechInc and FashInc, we found an extremely skewed distribution of customer profitability — which was consistent with our expectations. To get a clear idea of the distribution of CLV scores, we rank-ordered all customers in descending order and aggregated customers into 10 groups; each group represented 10% of the customer base. For ease of analysis, we then divided the customer base of both companies into three segments: High CLV (corresponding, at both TechInc and FashInc, to the top 20% of customers), Medium to Low CLV (corresponding to the middle 60% and the middle 50% of customers for TechInc and FashInc, respectively) and Negative CLV (corresponding to the bottom 20% and the bottom 30% of customers for TechInc and FashInc, respectively). The average CLV of one of TechInc’s High CLV customers was about 25 times as much as that of one of the company’s Medium to Low CLV customers. Interestingly, at both companies, the top 20% of customers contributed more than 90% of the company’s profits, because each company also had a sizable proportion of customers on which it lost money.
The skewed customer profitability distribution meant the companies should apply different marketing strategies to customers in the High CLV segment than to those in Medium to Low or Negative CLV segments. A direct implication of this was a paradigm shift in the allocation of the marketing budget. Previously, both companies had allocated their marketing budgets based on managerial rules of thumb as well as past performance of the overall customer base. Now, the marketing budget was allocated based on the projected future value of the customers, as indicated by each customer’s lifetime value. Both companies reallocated a major portion of their marketing budget from Medium to Low and Negative CLV customers to High CLV customers and then redesigned their marketing strategies in order to differentiate on the basis of the projected three-year lifetime value of the customer.
Higher Customer Lifetime Value,
Higher Stock Prices
For example, TechInc redesigned its acquisition strategy by focusing customer acquisition resources on prospects that closely matched the profile of existing High CLV customers, instead of targeting every customer on the prospect list. FashInc redesigned its retention strategies by selectively choosing High CLV customers for proactive relationship-building initiatives such as special rewards and shopping privileges. FashInc also differentiated its promotion strategies by offering High CLV customers unconditional incentives to purchase a new product category, while some Medium to Low CLV customers and all Negative CLV customers were offered such cross-buying incentives only contingent on spending a minimum of a certain amount. Both TechInc and FashInc redesigned their channel strategy by offering special incentives to the High CLV and the Medium to Low CLV customers to shop from more than one channel, while Negative CLV customers were encouraged to interact only through the online channel for both purchases and customer service. Several other marketing strategies were similarly redesigned, differentiated and implemented based on the lifetime value of the customers involved. The differentiated marketing strategies were directed toward increasing the projected lifetime value of the customers or lowering the cash-flow-related risks by reducing the volatility and vulnerability of cash flow streams from customers.
We measured the net outcome in terms of the overall lift in the lifetime value of all customers of each company. The results indicated that using the CLV metric to allocate the marketing budget and differentiate the marketing strategy helped both companies maximize their return on their marketing spending. In particular, both companies’ High CLV customers showed the strongest percentage increase in lifetime value. Failure to differentiate marketing budgets and strategies would have led to wasting marketing resources on Medium to Low and Negative CLV customers who had an inherently low potential for increased CLV.
Then, we applied the link function to predict the lift in stock price (or market capitalization) of the company based on the increase in the lifetime value of its customers. We repeated this procedure to predict the stock price of the company for nine months from the time of implementation of the marketing strategies. We then compared the predicted values with the actual stock prices and found that during the nine-month observation period, the marketing strategy outcomes (as indicated by the increase in CLV) corresponded to the actual stock price movement for both companies within an error range of 12% to 13%. An important implication of this result was that the marketing department was actually able to quantify the increase in stock price based on the performance outcomes of marketing strategies.
Furthermore, nine months after implementing the strategies to increase CLV, we found that while TechInc’s stock had gone up by 33%, its top three competitors’ stock had increased by an average of 12% over the same period. Similarly, while FashInc’s stock price had increased by 58%, the stock price of its top three competitors had risen only 15% on an average. (See “Higher Customer Lifetime Value, Higher Stock Prices.”) The stock price movement of both TechInc and FashInc also substantially outperformed the Standard & Poor’s 500 index (in terms of percentage increases) over the time period studied.
In summary, implementation of marketing strategies based on CLV enabled the marketing organization in each of the two companies studied not only to outperform the competition but also to create shareholder value. Conventional wisdom dictates that marketing is an art, that the marketing function is a cost center and that the marketer is a tactician primarily concerned with creative areas such as image and branding. In reality, the current business environment drives marketing to be more of a science than an art, the marketing function to be
a critical profit center and the marketer to be a strategist capable of increasing the company’s market value. The CLV framework we describe offers businesses the means to achieve this transformation successfully.