|
Topics and Objectives:
Recency (R), Frequency (F), and Monetary (M) have been called the "Three Magic Words" and for good reason.
Marketers in general, and database marketers in particular, have discovered that past behavior is predictive of future behavior. Customers who have purchased recently, frequently and have spent large amounts are more likely to be 'good' customers in the future than infrequent customers from the past who never spent much. The RFM approach to segmenting customers into groups is intuitive, simple to do, and relies on customer data commonly available in a company's database.
This module explains how to do RFM analysis and how it can be used to predict which groups of customers are likely to respond (or not likely to
respond) to a particular marketing campaign. RFM analysis is the first and simplest predictive modeling approach that we will consider.
Required Readings:
Recency, Frequency and Monetary (RFM) Analysis
This note 'walks' you through an RFM analysis using data on customers from a specialized mail order book club.
Here RFM is used in a test to see which RFM cells of customers are most likely to respond to BookBinders latest offering. The results can then be used to predict how the remaining customers will respond and BookBinders can 'target' likely responders rather than use a mass mailing approach.
RFM: Is it "Kudzu" or Is it Gold?
RFM is simple, intuitive and works surprisingly well. However, more powerful and sophisticated approaches are available.
Is it time to abandon RFM in favor of more sophisticated approaches - or does RFM still have a place in the toolkit?
RFM Migration Analysis: A New Approach to a Proven Technique
A short article discussing a different use of RFM analysis by FedEx. Fedex uses the three basic components - but has their own formula for
combining them into an RFM score. This is then tracked over time to see which customers are becoming more (or less) frequent, spending more (or less), etc.
Back to top
|
