Professional Documents
Culture Documents
RFM Response Rates
RFM Response Rates
RFM Response Rates
Balan Auhor: Carmen Balan Institution: Academy of Economic Studies of Bucharest, Faculty of Marketing, Department of Marketing, Address: Piata Romana no. 6, Sector 1, Bucharest, Romania, Phone: 021 319 19 80; 021 319 19 00 / ext. 469, 237, E-mail: cbalan@ase.ro The paper focuses on customer relationship management (CRM) underlining its analytical contribution to the substantiation of effective marketing strategies and programs. More specifically, the objective of the paper consists in the presentation of the RFM method, one of the most widely used tools in analytical CRM, in order to segment the customer portfolio of the company and identify the best targets for a particular offer or campaign. The acronym RFM corresponds to the following terms: recency, frequency and monetary value. Two alternative ways to apply the RFM method are analyzed. The former is based on sorting customer data in equal quintiles and analyzing the resulting data. The latter is centered on the calculation of relative weights for the recency, frequency and monetary value on the basis of the regression technique. The method is based on data that are available internally, being acquired by companies on continuous basis about their customers. The databases represent a real gold nugget for the companies that are interested to improve their performance by generating value for both customers and shareholders. Key words: customer relationship management, recency, frequency, monetary value, segmentation At present, CRM is one of the acronyms that are widely used by business and marketing managers, as well as by IT specialists. In essence, the abbreviation corresponds to the terms customer relationship management. Most market players who are or would like to enter the first league have designed CRM strategies and have invested in dedicated software solutions, in order to increase customer retention, boost organizational profitability and acquire sustainable competitive advantage.
766
767
Table 1 The hypothetical distribution of response rates by quintiles Recency Quintile Q1 Q2 Q3 Q4 Q5 Average Response rate (%) 5.2 4.6 3.1 1.3 0.8 3.0 Frequency Quintile Q1 Q2 Q3 Q4 Q5 Average Response rate (%) 4.1 3.8 3.2 2.0 1.9 3.0 Monetary value Quintile Q1 Q2 Q3 Q4 Q5 Average Response rate (%) 3.8 3.4 3.2 2.4 2.2 3.0
After the coding of each quintile, marketers are able to associate a set of codes with each customer in accordance with their specific recency, frequency and monetary value. For instance, a customer may be associated the figure 1 according to purchase recency, 2 in terms of frequency and 5 in terms of monetary value. In fact, the customers of a specific purchase recency quintile may belong to one of the quintiles 1 to 5 in terms of purchase frequency and to one of the quintiles 1 to 5 in terms of monetary value. The next step consists in the calculation of the breakeven value (BE). It is calculated using the formula presented bellow:
BE =
(1)
where UCP is the unit cost price and UNP is the unit net profit. Assuming that the cost of the mail sent to each customer is one Euro and the net profit is 50 Euros, the breakeven value is 2%, respectively (1Euro/50 Euros)*100. The value BE = 1 shows that the marketing campaign did not generate any profits and it only broke even. A value higher than one (BE > 1) reflects positive results achieved by the campaign. The breakeven value may be calculated for each customer and compared to the response rate achieved in the mailing campaign. The attractiveness of each customer account is reflected by the breakeven index. The indicator is computed using the following formula:
BEI =
RR BE 100 BE
(2)
where RR is the response rate to the marketing campaign (percentage of customers that have responded positively to the marketing campaign), BE is the breakeven value. For example, if the response rate is 3% and the breakeven is 2%, the value of the breakeven index is [(3% 2.0%)/2.0%] * 100 = 50. The values BEI > 0 indicate that profits have been generated by transactions. The value BE = 0 corresponds to a break even situation. The values BE < 0 indicate that transactions ended with losses, not with profits. Marketers should target only the customers with a positive breakeven value index. The customers with negative values of the BEI should not be considered for that specific marketing campaign because they do not generate any profit. The company must target only the profitable customers. It will gain more by focusing on the best customers than by mailing to all the customers in the database. A hypothetical situation is presented in the table bellow.
768
Table 2 Comparative analysis of the potential profits Indicators Average response rate Number of responses Average net profit/sale Net revenue Number of mails sent Cost per mail Total mailing cost Profits M.U. % Test 3 1,800 50 90,000 60,000 1 60,000 30,000 Full customer base 3 18,000 50 900,000 600,000 1 600,000 300,000 Target segment based on RFM 20 7,600 50 380,000 38,000 1 38,000 342,000
The data from the above table shows that a campaign targeting the most attractive customers leads to higher profits than a campaign directed to the full customer base.
where S is the RFM score corresponding to a specific customer, Pr is the number of points associated to the customer according to the purchase recency, Wr is the weight of the recency variable, Pf is the number of points associated to the customer according to the purchase frequency, Wf is the weight of the frequency variable, Pm is the number of points associated to the customer according to the monetary value of the purchases made by in a specific period, Wm is the weight of the monetary value variable. For each purchase of each customer, the recency score is computed based on a specific rule. A possible example is the following: 20 points are allocated if the purchase was made 3 months ago or less than 3 months ago, 10 points if the purchase was made in the interval (3,6] months ago, 5 points for (6, 9] months ago, 3 points for (9, 12] months ago and 1 point for more than 12 months ago. Similarly, the frequency score for each purchase may be calculated multiplying the number of purchases within the last 24 months by 4 points, but no more than maximum 20 points to each purchase. The score for the monetary value is calculated as one tenth (10%) of the purchase value, up to a maximum of 20 points per purchase. The regression will be used to identify the weights of the three variables (R, F, M). An example of possible weights is the following: 5 for recency, 3 for frequency and 2 for monetary value. The next table presents a hypothetical case of three customers.
769
Table 3 RFM score calculation Total score per customer 209 A B C C C 2 1 1 2 3 7 13 2 4 6 2 1 3 3 3 80 50 60 70 120 5 1 20 10 10 8 4 12 12 12 8 5 6 7 12 65 27 148 100 110 358 27 Frequency (number of purchases in the last 24 months) Monetary value score 10 Frequency score Customer Monetary value (Euros) Total weighted score per purchase 144 Recency (months) Purchase Recency score 20
100
According to the total score corresponding to each customer, the most attractive account is C (358 points), followed by A (209 points). Customer B (27 points) will not be considered as target for the marketing campaign due to the very low RFM score. The RFM method is a very useful tool for marketers in the process of customer target selection. Both alternative ways of application may be considered by companies. With the support of IT solutions, the RFM may be used on continuous basis for the assessment of customer attractiveness. The monitoring and evaluation of the recency, frequency and monetary value of the purchases facilitate the efficient segmentation and targeting of the customers existing in the portfolio of the company and the design of the most effective strategies in customer relationship management.
BIBLIOGRAPHY
1. 2. 3. 4. Payne A.F.T., Frow P. A Strategic Framework for Customer Relationship Management, in Journal of Marketing, October, 2005. Peelen, E. Customer Relationship Management, Pearson Education Limited, Prentice Hall, 2005, p. 4. Treacy M., Wiersema F. The Discipline of Market Leaders, Harper Collins, New York, 1996. Spiller L., Baier M. Contemporary Direct Marketing, Pearson Education, Upper Saddle River, New Jersey, 2005.
770