Granted, I have a smart dog, a young poodle named Charley. It usually takes only a few days and a few treats for Charley to master a new skill. Many traditional marketers have a harder time with something new, and we know about the difficulties of teaching old dogs new tricks.
I’m talking about marketers learning alternatives to RFM (Recency-Frequency-Monetary value), an old-fashioned but popular method for scoring customer behaviors. Just today I saw two more organizations still leaning on this outdated crutch.

First, there is the latest post on MediaPost’s Email Insiders blog, “Embrace the unknown. Then embrace it again.” In an otherwise good section titled, “Cohort performance, not campaign performance,” the author urges measurement by looking at “changes in RFM (Recency, Frequency, Monetary Value) scores over a period of time.”

Second, in the latest webinar from IBM on Marketing Analytics, “Which Metrics Matter? Three Things You Need to Know about Analytics,” again the recommended methodology is RFM. Folks, this is IBM, which recently acquired Unica and SPSS. You’d think IBM would know better.

What’s wrong with RFM

We are less than enthusiastic about RFM for two reasons:

1. RFM has no product information in its scoring. Cross-sell is the best way to grow revenue, but doing it right requires knowing what products each individual customer might buy for the first time. With no product information, RFM can’t predict purchase propensities for this incredibly important marketing technique. Our white paper, “How to Grow Revenue with Cross-sell,” lays this argument out in detail.
2. There are newer, better methodologies available!

So why is RFM still hanging around? We suspect it’s because some marketers haven’t looked at the alternatives recently. So c’mon IBM and MediaPost and any other company clinging to RFM.  Let’s have a shoot-out at the marketing corral.  We could even consider letting Charley be the judge. He knows almost as much about marketing as any company still relying on RFM.