Customer lifetime value (CLV) is an iterative process of maximizing net profit at the individual customer level. CLV is increasingly viewed as an important measure of the long-term health of a business based on the strength and understanding of the customer relationship. At a tactical level, CLV can give us an indicator of the marketing spend for new customer acquisition, while helping us expand upon the strength of our existing customer relationships.
Using predictive analytics we maximize CLV-driven profits by identifying triggers over the lifetime of a customer relationship. We define discrete stages of a customer’s financial maturity. We call these stages CLV mileposts.
CLV mileposts are pivotal moments that signify the approach of a changing need. CLV mileposts can be used as triggers that signal when customers may begin to default on a particular loan or when customers may begin shopping for their next financial product because of a real or anticipated financial windfall.
Identification of the CLV milepost structure allows us to predict individual customer behavior based on the behavior of other similar customers. The identification of these mileposts is usually a result of collaboration between the banking subject-matter expert and the analytics team.
Traditional customer models are based on generic data and only tell us who the (generic) “best” customer should be. Such models cannot predict how customers are going to change and/or evolve in the future. Using predictive analytics, we have a more powerful approach at our disposal.
Customer lifetime value is quantified in two ways. First, as a measure of a current customer’s potential, and second as a measure of dollar for dollar profitability to the bank. Each customer interaction has the potential to increase or decrease CLV depending on the customer’s perception, the fitment of financial products offered and the timing of those products relative to the other customers within the bank. CLV done properly is more than offering financial solutions. CLV is a balancing act of offering the best product at the most appropriate time, to the most profitable customer.
CLV done properly is performed at the level of the institution. Therefore, instead of simply modeling individual customers, we ‘roll up’ the customer model to the environment level of the business and determine how to maximize each customer contribution within that environment. Customer potential using CLV is a measure of individual spend and is optimized as the services of the bank evolve. CLV is designed to cater to new and existing customers that are currently seeking services at a real or anticipated point of need.
In order to identify spend potential, CLV relies on mileposts. CLV mileposts are a moving target that evolve over time. We must extend the identification of CLV mileposts to relative stages within customer segments. Probability models are built allowing mileposts to be flexible and meaningful for each customer.
The idea of modeling CLV mileposts is intuitive. By determining the path of the evolution of a customer relationship, we can identify the current milepost for a particular customer and predict the timing of the customer’s transition to the next milepost. We then model the transition for financial opportunity and potential risk, in tandem.
Modeling CLV mileposts gives tangible results that are analogous to being hooked up to a “heart monitor” in which the portfolio is not just checked for current opportunities but is consistently monitored for new and existing potential opportunities.
Monitoring customer behavior is a continual and iterative process. By monitoring behavior patterns, corrective action can be applied to remedy the path of a diverting process before it impacts profitability. By identifying when a CLV milepost event is most likely to occur and the affecting attributes, we allow ourselves the opportunity to adapt. The result, happier customers that get superior products and services at the point of need. We now gain the ability to increase wallet share and manage risk at the same time.
The efficient and intelligent use of predictive analytics for predicting customer behavior drastically changes the banking experience not only for the bank but the customer as well. Servicing the needs of your customers by anticipating those needs before they happen brings a level of fluidity between yourself and the ever-changing needs of a diverse customer portfolio.
CLV begins with segmentation of the customer base into meaningful groups based on customer data. Segmentation is designed to uncover hidden attributes or factors about customers that cannot be measured by or within any single characteristic or trait. Segmentation considers all customer attributes and tells a story of ‘why’ customers are different as a result of all of the information available in the data. The process of CLV then identifies CLV mileposts within each customer segment, while providing the recipe for identifying attributes and economic conditions about customers that are in need of new services, or restructuring of current services. By employing CLV, every customer is potentially profitable depending on the terms and timing of products and services offered.
The identification of CLV mileposts is an estimated mapping of the progression of current customers as they financially mature within the portfolio. The identification of CLV mileposts give a clear description of customers within each segment. We have a description of the single best customer along with a description of all of the ‘best’ customers, as they relate to each milepost. We can now target specific profiles for specific services and even bundle services together as the customer progresses. Additionally, since mileposts are fluid, CLV can tell you how those same customers are likely to react in the future.
In addition to the aforementioned, CLV also provides a powerful tool in ‘time-to-event’. By accurately estimating time-to-event, we can proactively anticipate and manage the changing needs of your diverse customer portfolio.