RFM Analysis


An RFM analysis is a scoring process with which you can sort customers into target groups or segments using different KPIs. It aims to identify customers who are most likely to respond to a range of different marketing methods. RFM analyses are most frequently used in direct marketing, but also find application in email marketing, campaign optimization, performance and database marketing. Based on the scoring values ​​recency, frequency and monetary, customers are divided into buyer groups in order to find out which customer groups are particularly profitable and with which type of customers certain campaigns or marketing offers are less worthwhile. The purpose of RFM analysis is to forecast campaign return rates and increase return on investment.

General information[edit]

The RFM analysis is an empirical method, which is strongly dependent on data from the web analysis, customer relationship management or transactions. While many marketing approaches are based on demographic characteristics, RFM analyses complement the strategic direction of campaigns with a behavioral component. To this end, the purchasing behavior in the past is examined in more detail:

  • R - Recency: The recency of a purchase is an important tool to identify customers who have bought something recently. Customers who purchased not too long ago are more likely to react to a new offer than customers whose purchase happened a long time ago. This is the most important factor in an RFM analysis.
  • F - Frequency: Frequency of purchase comes after recency. If customers have purchased more frequently, the probability of a positive response is higher than for customers who have only rarely bought something.
  • M - Monetary Value: The turnover of a purchase or the monetary value refers to all purchases made by a customer. Customers who have spent more money on purchases are more likely to respond to an offer than customers who have spent smaller amounts.

How an RFM analysis works[edit]

In practice, RFM analyses are usually carried out automatically. In extensive CRM systems, business intelligence software or predictive analysis systems, such functions are often already implemented. Scoring, which will be briefly explained below, is, of course, dependent on the available data.[1]

  • Based on the last purchase, customers will be assigned a score that indicates recency. Time intervals or the date of the last purchase can be used. The assignment of the recency score can be done with arbitrarily formed ratings. For example, five or less categories can be used to distinguish groups whose purchases are more or less recent.
  • With the same data, customers get assigned a frequency score that indicates how often customers have purchased in a specified time interval or since a certain date. Here too, the division into groups can take place generally at your discretion in order to serves the corporate objectives.
  • Next, customers are assigned a monetary value, whereby the highest level corresponds to the highest revenue from a customer group. For five categories, monetary values ​​can be selected from 1 to 5.

The actual RFM score is only completed now. Recency, frequency, and monetary scores are added or combined and yield the RFM scores for the respective customer groups. Overall, there are 125 possible RFM scores for five ratings, the highest possible combined score is 555. The highest scores will be assigned to the customers who most likely respond to an offer, be it a planned campaign or promotional actions. The likelihood of response is, however, only an assumption made based on the data.

In addition, different RFM scores can be generated. Depending on the database, the scores can result from transactions or aggregated customer data. Some programs also offer the visualization of the RFM score, for example, to display the customer groups as diagrams or scatter diagrams.

Practical relevance[edit]

The use of these findings in marketing is often viewed critically. While RFM scores can help identify the customers with the greatest purchasing power, marketing activities should not only address that customer group even if the bulk of sales are expected from it. Customer groups are arbitrarily divided. A type of profiling is used, which can be only poorly conveyed to the customers. If only paying customers are given special offers, there is a risk that other customers will feel discriminated against when they hear about it. Many marketing experts recommend to focus on the low-paying customer groups to increase their loyalty and purchasing power.[2]

Relevance to online marketing[edit]

RFM analyses can play a key role in increasing the profitability of advertising campaigns. Moreover, RFM analyses can be used in different channels and for various purposes such as the response rates to email campaigns, the transactions in an online shop, or the conversions of websites which offer a whitepaper for download. Campaign optimization, segmentation, and deeper target group analyses can also be carried out based on RFM analyses. Example: Using RFM Analysis in Segmentation help.sap.com. Retrieved on December 02, 2016</ref> Most RFM models can be adapted to the respective conditions. However, they often also require a comprehensive CRM system that includes such analytical methods and a rather sophisticated data collection, which must be integrated and implemented using tracking methods. This is generally also possible in Excel, at least the transaction data has to be available.

References[edit]

  1. RFM Analysis ibm.com. Retrieved on December 02, 2016
  2. RFM analysis (recency, frequency, monetary) searchdatamanagement.techtarget.com. Retrieved on December 02, 2016