RFM model

RFM is a method used for analyzing customer value. It is commonly used in database marketing and direct marketing and has received particular attention in the casino and retail industries. RFM stands for:

  1. Recency—How recently did the customer purchase?
  2. Frequency—How often do they purchase?
  3. Monetary Value—How much do they spend?

 Most businesses will keep scores of data about a customer’s purchases. All that is needed is a table with the customer name, date of purchase and purchase value. One methodology is to assign a scale of 1 to 10, whereby 10 is the maximum value and to stipulate a formula by which the data suits the scale. For example, in a service based business like the casino business, you could have the following:

  1. Recency = 10—the number of months that have passed since the customer last purchased.
  2. Frequency = number of purchases in the last 12 months (maximum of 10).
  3. Monetary = value of the highest order from a given customer (benchmarked against $10k).

Alternatively, one can create categories for each attribute. For instance, the ‘Recency’ attribute might be broken into three categories: customers with purchases within the last 90 days; purchases between 91 and 365 days; and purchases longer than 365 days. Such categories may be arrived at by applying business rules, or using a data mining technique to find meaningful breaks.


Once each of the attributes has appropriate categories defined, segments are created from the intersection of the values. If there were three categories for each attribute, then the resulting matrix would have twenty-seven possible combinations (one well-known commercial approach uses five bins per attribute, which yields 125 segments).


Segments could also be collapsed into sub-segments, if the gradations appear too small to be useful. The resulting segments can be ordered from most valuable (highest recency, frequency, and value) to least valuable (lowest recency, frequency, and value). Identifying the most valuable RFM segments can capitalize on chance relationships in the data used for this analysis. For this reason, it is highly recommended that another set of data be used to validate the results of the RFM segmentation process.


Advocates of this technique point out that it has the virtue of simplicity: no specialized statistical software is required, and the results are readily understood by business people. In the absence of other targeting techniques, it can provide a lift in response rates for promotions.


Whichever approach is adopted, profiling will be done on the final results to determine what makes up group membership. Categorical factors such as gender, nationality/locality can be used as well as age (or, indeed, any other demographic feature that is available) to understand the “type” of customer that resides in each group. These factors can be used for each segment and applied against the population metrics to determine how much more or less likely a segment is to exhibit a particular feature or type of behavior when compared to the customer base as a whole.


A few words of caution in the gaming field: a major drawback of classical RFM modeling is the high propensity of a casino and/or a sports book to be continually hitting the same segment(s) with the same marketing message.