Life Events Approach: Strategies for Managing Customer Lifetime Value in Retail
Written By John Parkinson, Partner & Managing Director, ParkWood Advisors LLC
Edited By Mishu Rahman Din, Medill IMC Class Of 2015
Published on 10/21/2015
Customers have many choices when satisfying a retail need, and the exchange of value between a business and its customers can take many forms: from a single transaction to multiple interactions over a long period of time. However, in many retail situations, a customer’s needs and shopping preferences evolve over time, and unless the business adapts the relationship strategy, the customer will likely go elsewhere as alignment weakens or fails. In narrowly focused specialty retail, this may not be a problem as the customer cohort is constantly being replaced. But in other areas, such as department stores or when the retail chain segments its customers by age and aspires to serve them through multiple segment strategies, understanding what triggers a move between segments increases the chances that the business can retain and potentially increase the customer’s value.
A business will generally try to increase the lifetime value of a customer (which implies a relationship over time) along each of three dimensions:
- More spending within a category
- Spending in more categories
- A longer duration of engagement
Different investment strategies in each dimension can increase the yield and enhance lifetime value. The sum of these strategies can be thought of as a form of Customer Value Management (CVM). Efficient allocation of resources across the three dimensions can significantly improve customer lifetime value.
A “life event” model allows a business to monitor available information sources for data that indicates a trigger has occurred and that the value management system should adapt in response.
- Graduating from college
- Starting a job or getting a promotion
- Getting married or divorced
- Buying a new home or adding to an existing home
- Having a child
In the past, scanning for this trigger data was difficult and expensive, but the advent of the Internet and social media in particular, and the digitization of content in general, has made the process possible and cost effective. The accumulation of customer interaction data over an extended period of time also enhances the possibility of using predictive models to anticipate a consumer’s needs and to act proactively to capture the potential business. In many retail businesses, where margins can be very thin, capturing even a small increase in share of wallet can be critical to continuing profitable growth.
Some years ago we worked with a chain of department stores to build a prototype CVM platform using this life event strategy. The chain knew that its primary customer segment (accounting for more than 70 percent of their revenue) was women between the ages of 25 and 50 with a median income of around $40,000. Analyzing spending patterns showed us that although many customers shopped across categories within the store, they did not do so consistently. Survey data also indicated that customers often shopped elsewhere when they could have obtained better quality or prices in the chain’s stores. By tying together point of service, loyalty card and locally available public data (from newspapers and community sources), we built a predictive model of life events for a statistically significant proportion of customers. This allowed improved microtargeting for just-in-time advertising and marketing campaigns, reaching more customers at their most receptive moments and reducing the number of “lost” sales. Although the prototype was successful, the chain decided that the effort required to maintain customer histories rich enough to drive the predictive analytics would be too expensive and the platform was not deployed.
Later we re-implemented the same strategy at a specialty fashion retailer whose core product line was geared towards women, but were seeking to branch out to a younger audience with a new brand — and a new line of loungewear products that would be more appealing to teens. The new brand would only work as a CVM entry point if its customers could be retained as they moved from teenage to young adult and onwards, and the brands would have to carefully evolve their messaging and offers accordingly. Here the model used both point of sale and loyalty program data with a statistical model of customers’ ages at first contact, with analytics predicting the right time to suggest a switch from the juvenile to the young adult product ranges as well as additional purchases to account for individuals’ growth during adolescence. In-store and post-purchase survey data were used to improve the initial predictive models. This data surfaced a younger than anticipated customer segment that required us to extend the period before proposing a switch to a more adult range of products by over a year. It also identified an opportunity for a broader product range within the brand. The successful prototype was eventually incorporated into the brand’s overall OmniChannel marketing and positioning strategy and continues to be an effective customer retention tool.
Finally, we worked with a luxury retail chain to identify and understand the evolution of their core of most profitable customers. The chain depended on a relatively small proportion of customers for roughly half of its revenues, yet did not understand very well where customers initially appeared from, how customer preferences and spending patterns changed over time, or why some “regular” customers suddenly disappeared. The chain also had three sales channels (store, online and catalog) and needed to better decide how to invest in each to maintain and grow their most profitable customer segments. We were able to build a CVM platform that integrated with their concierge (personal shopper) support technology and better identify and target the evolving behaviors of highvalue customers across all three channels, significantly reducing “churn.” Better understanding of this core segment’s lifetime value also allowed the retailer to refocus advertising and promotional investment (away from less profitable segments, where the impact did not justify the expense) towards programs that increased the level and frequency of “touch” with their best customers.
As evolving technology and social behaviors create more and more opportunities to accumulate information about customers, the potential for customer lifetime value management will continue to increase. Retailers that understand their customers’ evolving needs will be able to build mutually beneficial long-term relationships. Those that do not will be relegated to merely transactional commerce. [END]