Measure What Matters… Critique of the RFM Model

I’m always interested in new strategies that could deliver deeper insight into our customers and would-be customers. But since burying my head into the mountains of marketing and management resources available, I’ve noticed a danger that managers can often fall into… Paralysis-by-Analysis. This is the tendency to crunch and re-crunch the numbers from every angle before deciding to crunch them again, and if there’s time, take a course of action.

Good marketers will bind themselves to ACTIONABLE analysis only. Whether it be customer survey data, customer behavior data, industry trend data, competitive analysis, or any other analysis activity, let us keep this question on our minds, “How will this information help me market MORE accurately and effectively.

I put the emphasis on “MORE” because you can waste your time tracking metrics that tell you something you already know from an ACTIONABILITY standpoint. Even if some new insight into your customers can enhance your marketing efforts, it needs to be approached from a cost/benefit perspective. In other words, “Will the changes I make to my marketing program from this new insight produce enough ADDED profit to justify the resources (time, energy, money) it takes to retrieve that insight?”

I asked myself this same question when looking into the Reach/Frequency/Monetary (RFM ) model that is used by many catalog retailers. Hats off to Mark Sakalosky over at Clickz.com, who wrote an excellent piece that educated me on the RFM Model. In a nutshell, RFM is a way of segmenting your customers according to how much they spend, how often they spend, and the last time they purchased from you.

Lets take a look at some possible scenarios with the RFM Model:

Goal (1.) Identify all the customers who rarely shop but purchase big ticket items when they do. Develop a marketing campaign targeted to increase the Frequency of purchases, essentially converting the customer to a more valuable RFM segment.

Goal (2.) Identify all customers who haven’t purchased in a while, but purchase big ticket items when they do. Develop a marketing campaign targeted to increase the Recency of purchases, essentially converting the customer to a more valuable RFM segment.

QUESTION: Is there a marketing campaign on this earth that will accomplish Goal (1) without simultaneously accomplishing Goal (2) ? This is just one way that the RFM Model seems to produce truly un-Actionable data.

As Marketers, we need data that can be used to form a marketing offer that appeals to the attributes of a specific customer segment. To be fair, some of the information delivered by the RFM Model can do this. And it makes more sense for catalog retailers, who heavily depend on the orders of their subscribers to implement the RFM Model.

But before you go through the pains of tracking, mining, and retrieving purchase behavior and customer information, consider segmenting your customers based on information that can help you form specific offers. Here are some good segments to track:

(1.) Order Value: One of the 3 elements of the RFM model, segment customers according to their average order value and give them an offer for products in their price range.

(2.) Product Category: Segment customers according to what they’re interested in and give it to ‘em.

(3.) Geographic Location: Offer an “Alabama Special Offer”.

Of course, there are many ways to segment your customers into meaningful groups. The key word here is “meaningful.”

Another problem with the RFM Model is that it reverses the customer-centric rule of marketing. Instead of discovering the desires of your customers and offering products to meet those desires, the RFM Model starts with your desire to get all of your customers to the “highest score” possible by increasing their Frequency, Recency, or Purchase Value. This can produce irrelevant offers that are out of touch with what your customers want.

Unless you work for a catalog retailer or have already implemented the RFM Model, I would focus on segmenting your customers where your Value Proposition meets their needs. This is the insight upon which high-performance marketing campaigns are built.

P.S. – I would love to hear other marketers’ opinions about the RFM Model.