Posted on April 25, 2012 · Posted in Uncategorized

RFM (Recency, Frequency, and Monetary Value) is typically used to measure buyer behavior. But can it be adapted to measure “browsing behavior” as well? Here is an example that will let you decide for yourself.

A membership association asked DMSI how they could measure web site use by members. Their theory was that active use by members of the association’s web site would indicate good targets for membership renewal. By requesting a member login with each visit, individual web behavior could be easily tracked. If Recency and Frequency are good measures of “ready to respond” in the physical world, why not in the virtual world?

I presented this concept to my Johns Hopkins graduate students as a potential research project. One student – Fran Atkinson – took up the challenge to study membership retention at the organization where she is employed (a large multi-disciplinary health association).
Ms. Atkinson created a “RVH” model (Recency, Visits, Hits) as a substitute for RFM. A sample of web site visitors over a 6-month period were scored using RVH, and the top 80 were selected for analysis. Demographic info (gender, years as a member, etc) was appended to the 80 records. The results of this study were interesting, although not completely conclusive due to the limited sample size:

  • High RVH scores did not correlate with a member’s actual registration for the association’s annual meeting.
  • High RVH scores did correlate with length of membership: Both new members and very veteran members (over 15 years) scored high in RVH.

From this information, the Membership Department agreed that web content needed to be changed to attract more traffic from individuals with between 3-15 years of membership. Another proxy for “M” in RFM could have been page views.

Professor's CornerProfessor’s Corner:

E-Tailing: Why RFM Still Matters

Originally, RFM was developed for cost-effective mail campaigns to customers that were likely to buy.  In a world where email is free, does it still make sense? The resounding answer from marketing experts is YES!

RFM analysis tells you who are your best customers – the 20% that are actively engaged with your products, services and organization.  It also classifies the remaining customers into 4 more tiered groups, again according to RFM scores.  Mailing to low RFM customers with the same frequency as high RFM customers may be free, but can result in “unsubscribes” and may get you listed with an ISP for Spam, blocking getting your content to ALL your customers!  Not exactly “free”.

Use RFM to sort and prioritize email campaigns.  High scorers will welcome frequent (but not too frequent) news and promotions.  Sent to your 2nd tier less frequently; consider the score gap in the next groups before sending emails at all.  Is there a slight or pronounced difference?  Large score differences may reflect customers who bought once and are not likely to return.  They are also the customers most likely to report too-frequent emails as Spam.

As always, call today to take advantage of our free 15 minute consulting session, “The Professor is in” at 703.941.8109.