Do You Really Know the Customers in Your Loyalty Program?

By Hilary DeCamp, Chief Methodologist at Material

Accurate, timely data collected directly from customers is a vital source of actionable insights to guide critical business decisions.

For brands, a primary benefit of offering loyalty programs is being able to harvest zero- and first-party data by capturing shopping behavior and applying targeted methods and tools. A secondary benefit is having easy access to validated customers to participate in surveys about service, new offerings and unmet needs.

But when the composition of the survey respondents does not mirror the overall loyalty program membership, or when the loyalty program membership itself isn’t reflective of the brand’s overall customer mix, the resulting data and insights are bound to be flawed.

We’ve seen this type of mismatch emerge more often than you might expect.

Let’s say high-value business travelers make up 50% of an airline’s frequent flyer program. Yet they account for only 25% of the respondents to a survey about preferred perks, with members who fly for personal trips only once or twice a year making up the majority of the respondents. Were the airline to develop new perks based on the responses to this particular survey, they would likely please the lower-value infrequent travelers but could cause the more-profitable higher-value frequent flyers to disengage, hurting the airline’s top and bottom lines.

Fortunately, brands can take steps to avoid such costly mistakes. First businesses must segment their membership data so they can accurately compare the makeup of survey respondents to that of the audience overall. They would ideally classify their customer database members by RFM (recency, frequency and monetary value) or a similar framework to be sure the response to a survey does not over- or under-index significantly in terms of engagement, value or other critical criteria.

From there, brands need to understand the key reasons behind these discrepancies before they can correct them.

Battling Nonresponse Bias

Nonresponse bias – also known as participation bias or contact and cooperation bias – occurs when certain segments within an audience are more or less likely to respond to a survey than other segments. In the example of the frequent flyer members, on-the-go business travelers might be too busy and therefore less inclined than leisure travelers to participate in surveys. There are a variety of ways to combat nonresponse bias.

• One way to mitigate nonresponse bias is weighting: giving more weight to the answers of people who belong to underrepresented groups and less weight to those within overrepresented groups.

• Another option is response-balanced outgo, inviting a disproportionate number of members from the groups least likely to respond. To give a simple example, if a loyalty program’s database is evenly divided between men and women — but in the past women were twice as likely as men to respond to surveys — a brand might send the surveys to twice as many men as women. This solution requires a sizable database to draw upon.

• Relying on incentives to boost cooperation can be tricky. If the incentive is a discount on the brand’s product or services, the respondents will skew toward fans and deal-prone individuals. Depending on the objectives of your survey, this bias may be acceptable. But if it’s not, then more generic incentives would be worth their higher cost.

Overcoming Attitudinal Bias

Even after correcting for nonresponse bias, it’s still common to find that attitudes among loyalty program members differ – sometimes markedly – from those of a brand’s total customer base. This is most apt to occur when the only clear benefits of joining the program are deals and discounts. Deal-driven consumers may well make up the overwhelming majority of members even if they’re only a small portion of your overall customer base. In your RFM matrix, the loyalty program members might score higher than average in frequency but lower in average order size — and are likely less profitable to boot.

To ensure your loyalty program is a robust zero- and first-party data asset that helps you better understand customers and drive effective personalized offers, structure it so the rewards are not limited to cost savings. For example, a beauty brand promoting eco-friendly product lines could offer to donate a portion of proceeds from members’ purchases to environmental causes; a footwear retailer could provide loyalty club members early access to limited-edition drops; a concert venue could enable program members to earn points toward listening parties or pre-show autograph sessions.

When Material helped Dr Pepper revamp its Perks program, we dug deeply into customer motivations. Among other changes, we introduced rewards such as member-only access to new flavors. The new Pepper Perks program generated a 56% increase in acquisition along with improved ROI.

Deeper Customer Understanding Yields Greater Business Impact

A recent IAB survey found that 71% of businesses are working to increase their first-party datasets, up from 41% two years prior. If that data comes from consumers who do not accurately reflect a business’s broader audience, the actions taken by the organization could actually end up hurting its performance, particularly by leading to bad pricing or promotion decisions.

At Material, we have decades of experience helping businesses build and optimize loyalty programs that not only boost short-term performance but also provide reliable, actionable data and deep human insights to fuel long-term success. Get in touch with our experts today to learn how we can deepen your customer understanding and make a measurable impact for your brand.