When the Obvious Isn’t So Obvious

I recently read about a study published by Cardlytics that summarized "a 'whole-wallet' analysis of transaction records held by banks for nearly 70 percent of U.S. households." Sounds like it should contain some meaningful insights – we've often said that if you're looking at transactions by your customers, you're getting an extremely limited view of their shopping behavior, only what they do at your store, not within your category or in other categories, meaning you need to draw broad conclusions carefully. So I read with interest. The key takeaway:

The research indicates that customers who frequently visit specific retailers tend to be heavy "category spenders", meaning that they also frequently visit other retailers in the same channel. Instead, true loyalty is often the domain of "light customers" who make fewer trips to stores but typically shop at the same ones.

Not with-standing the argument that frequency is a behavior and loyalty is an emotion, this seems fairly obvious: less frequent shoppers have fewer opportunities to make alternate choices, so they are likely to focus their spending at a familiar location. But is it really so obvious that this means they are more loyal?

The study considered shopping behavior in five common categories: restaurants, apparel, gas and convenience, grocery, and general retail. The core of the analysis was an evaluation of how often consumers visit a single store in the context of their overall visits in the category. The definition of a "loyal" customer was one that concentrated more than 50% of their visits at a single retailer. Behavior within a category was considered as a basis for comparison across categories.

That's the first issue that I see with the methodology. Motivators and decision factors in each of these categories can be wildly different. For example, within restaurants, need for variety is a consideration that would never apply in gas and convenience or general retail. Frequency of visits across all of the categories would vary widely, and convenience as a driving factor would be substantial for gas and grocery, less so for apparel. So you would have to consider each in context. The second issue is with how they seemingly categorize light and heavy buyers – light are defined as those with less than 52 trips a year, and heavy as those with more than 130 trips a year. Yikes, that sounds like a lot of mall visits to buy clothing. So back to my recent comments on expectations, you might want to start by thinking about how we expect consumers to behave within those categories.

Nevertheless, the study concludes that there are significant differences in loyalty across categories: restaurant customers are the least loyal, with 44% of visits to a single retailer, and grocery customers are the most loyal, with 81% of visits to the same retailer. I'm not sure I agree – if a restaurant gets 40% of a consumer's total category visits, that seems like it may be a pretty loyal customer. American Eagle Outfitters might expect to get a higher degree of spend from a teen than Gap would expect to get from a young adult. And a grocer might be concerned that it is only getting 80% of category visits – what is missing from its offering that it can't meet all of a consumer's needs?

Many of expert opinions in reaction to the article ran along the lines that retailers are probably more focused on the frequent shopper anyways, so this definition of loyalty is suspect. The obvious is obvious, but also irrelevant. That mostly misses the point. This findings from this study actually highlight the need for a retailer to go deeper with their analysis, to consider frequency and concentration in order to identify opportunity within each frequency tier. By doing this you can develop tailored treatment plans that focus on growing long-term value within all of the frequency tiers, not just the most frequent.

Posted by jkeenan on 04/10