In my earlier post about customer satisfaction assessment practices, I shared information about the Apostle Model and how it can be a useful tool for segmenting your customers by loyalty profile. Today, I’d like to share an additional model which also has a four quadrant structure like the Apostle Model, but which is focused on understanding the value and growth potential of specific customer segments.

This model was cited in the Harvard Business Review by Werner Reinartz and V.Kumar in their paper entitled The Mismanagement of Customer Loyalty. (download the full paper as a PDF by clicking the link). One of the big takeaways from this paper is that loyalty does not necessarily equate to profitability. In fact, their research shows that companies have a tendency to keep investing in customers who may not be loyal now but who were loyal in the past, or who purchase at unprofitable levels.

It also appears that long-term customers don’t cost less to serve, are not willing to pay more for products, and don’t necessarily promote your brand more. So, the key to loyalty programs is to segment the market so that you invest in only the customers that are loyal as well as profitable and outspoken about your brand. The latter point is probably the most important because brand advocacy is the area where loyal customers can have the greatest impact. The researchers propose a model to segment customers, thus assisting the direction of resources, which I’ve interpreted below (you can see their version in their paper). You can view a larger version by clicking on the image below:

loyaltystrategymodel

True Friends tend to be satisfied with their relationships with companies. For them, managing programs is about maximizing returns without going too far. Offer overload can lead them to start ignoring communications all together. Butterflies on the other hand are easy come easy go, so you have to get the most out of these relationships while you can with promotions or other incentives. But, make sure to stop investing in Butterflies as soon as they leave, because any extra spending will cut into your profits. Barnacles are the most challenging, but they do hold potential. The key is determining if they’re just hanging around for low-profit deals, or if they have the potential to buy different and upmarket products. If the latter is true, then carefully crafted loyalty programs may help, but make sure to have strict controls in place to make sure you don’t spend more on loyalty than they offer in profit. Finally, with strangers you just have to make sure that each transaction comes with some profit.

With the above chart it is possible to define a strategy for investing in loyalty programs based on these four specific segments, but I do want to add a final note about how to assess customer loyalty. The researchers talk about an over reliance on the use of recency, frequency, and monetary value (RFM) calculations which can produce misleading results. I don’t want to dig into this topic here, but the key is to base loyalty assesments on each individuals’ buying patterns, rather than on community standards or simple weighted additive RFM measures. This is because buying patterns vary widely and could cause you to interpret a True Friend as a Butterfly. Check out the research paper for a more detailed explaination, or consider talking with a predictive analytics company about how their model works.


The first time you leave a comment, it will be placed in a moderation queue. Once we know you are not a bot, you will be free to comment at will from then on.

Leave a Reply