Implementation of RFM Analysis in Determining Customer Relationship Management Strategies for Beauty Care Services

In the health services industry, beauty clinic is one of the areas of the business which developed enough at the moment. Not only is synonymous with women, the men also love to come to these places to improve the appearance. This study adopts the techniques of data mining, namely SOM and K-means that the objective data grouped into one or more groups form, which is then applied to the analysis model into RFM (recency, frequency and monetary) at one of the clinics beauty in Jakarta. The result is a model of analysis of the RFM was able to produce a grouping of customers and can increase the accuracy of the information behaviour of beauty clinic customers. RFM analysis models help identify four types of customers, including the loyal customers, potential customers, new customers and lost customers. RFM analysis results showed that cluster 1, namely “new customer” has a value of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$R$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$F$</tex> is high but the value <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$M$</tex> is less than the average of the RFM. Cluster 2, namely “loyal customers” have a value higher than the RFM on average value of RFM. Cluster 3, namely “Lost” has a low of RFM on average value of RFM. Cluster 4, namely “potential customers” have a value of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$R$</tex> and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$M$</tex> is high but the <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$F$</tex> value is lower than the average rating on RFM.