Loyalty programs that reward customers for repeated and incremental spending are popular with consumers and brands—and for good reason. Consumers benefit financially and emotionally from belonging to programs that reward them for engagement, and brands secure higher spending, incremental revenue, and reduced retention costs associated with loyal customers. Done correctly, loyalty programs are a win for everyone involved.
They’re a considerable investment, however. The variable cost of the currency redeemed (e.g., points, miles, etc.), fixed program administration expenses, and the necessary investment in tech infrastructure to create an app and a website to support the initiative all combine into a significant expense.
Given the costs, how does a brand know if its investment will generate a positive return? Estimating the financial impact of a loyalty program is challenging, but data can help a brand gauge if a program will positively impact the business and if it will remain financially viable for the long term. Let’s explore best practices and important considerations.
Customer retention isn’t the only goal of loyalty marketing programs, but it’s often the primary one. Quantifying the actual value of keeping customers, however, can be difficult. With so many personal, logistical, and financial factors influencing purchase decisions, how can we isolate and measure a loyalty program’s impact?
First, we need to answer two core analytic questions: How many customers would be lost, and/or transact less, if the loyalty program didn’t exist? And what would that represent in lost revenue?
The usual best practices may not be practical
In an ideal world, standard scientific methods would determine a program’s return on investment (ROI). These two analytical approaches are the usual ways to analyze the probable financial return of a proposed loyalty program.
Comparing member behavior with a randomly selected control group
The most effective and precise way to measure the impact of any strategy on human behavior is to compare the behaviors of a group of people exposed to the strategy to those of a randomly selected control group who aren’t exposed to the strategy. The difference in their performance measurements illustrates the strategy’s impact.
This approach isn’t practical for most brands. Implementing a control group for a loyalty program would require allowing some customers to join the program and excluding others for the sake of measurement. That’s impractical, of course, because loyalty programs are typically quite visible—i.e., they’re administered on public-facing apps, on websites, and in stores. In addition to the potential legal and public relations issues with excluding some customers for the purpose of measuring ROI, the effort and expense required to prevent certain customers from joining a program are too weighty compared to any benefits provided by the collected data.
Another avenue that would lead to similar results might be a pilot program launched in select markets to measure results from similar groups. But sustaining that for extended periods of time also isn’t practical, especially in industries with low transaction frequencies (e.g., automotive, travel, and entertainment). For low-transaction industries, estimating the financial impact of a loyalty program requires years of transactional data. It’s difficult to imagine maintaining a loyalty program pilot for years.
Comparing member with nonmember behavior
Without a control group, the next-best approach is to compare behaviors of program members to those of nonmembers. This approach also has a major drawback—in short, self-selection bias. Customers who opt in to a loyalty program are more likely to be repeat purchasers and more engaged with the brand than those who don’t voluntarily join a program, and that can skew data.
To account for self-selection bias, we recommend developing an enrollment propensity model. Enrollment models can be used to ensure the members and nonmembers analyzed are equivalent in terms of joining the loyalty program. Executing an enrollment propensity model requires an investment in time and resources that may not be feasible for every brand, especially those without a dedicated analytics team. That said, an experienced team—with expertise in this type of modeling, transactional data, and demographic data—can identify nonmembers exhibiting traits similar to members and compare the results of members and nonmembers with the same likelihood of enrollment.
If the best scientific methods aren’t reasonable options for most brands, then how can they measure the financial impact of a loyalty program? Consider these suggestions:
Measure impact within unique business units. One practical way to measure the financial impact of a loyalty program is to consider its effect on specific business units. Most businesses generate revenue from a variety of channels. A car manufacturer, for example, earns money by selling vehicles, but also by selling accessories and dealer services. Considering the value of incremental revenue generated by loyal customers in specific business units is a crucial element in measuring the value of a loyalty program.
Measure results within specific time periods. Economic and social factors—like the pandemic and the ensuing inflation—and even seasonal habits influence customer behavior. A business can account for the impact of timing on customer behaviors with the following approach: Identify members and nonmembers who were customers as of the same date. Next, within that group, measure retention and total transactions over the next 12 months. Thus, the period of measurement is the same for these member/nonmember cohorts, and the study can be expanded to include more than one year after enrollment (i.e., impact in years 2, 3, 4, etc.).
Report the impact results in terms of revenue and contribution margin. That’s how you can demonstrate how investment in a loyalty program is projected to affect business performance and results—which will help business leaders and marcom decision-makers contextualize the results.
Loyalty programs require significant resources to design, implement, and maintain. As with any investment, it’s important for a brand to measure and understand the impact that investment represents in customer retention, sales, and profits. Use these three suggestions to estimate how a loyalty program could pay off for your brand when more standard analytical approaches aren’t practical.
Also, keep in mind that part of your analysis should include assessing if your team has the experience and expertise to create and manage a program. You may need to seek out expert guidance for a specific part or to lead the way through the whole process.
Chris Veerkamp is a manager, Data Intelligence, for The Lacek Group, a Minneapolis-based data-driven loyalty, experience, and customer engagement agency that has been delivering personalization at scale for its world-class clients for more than 30 years. The Lacek Group is an Ogilvy company.