The RFM-FCM approach for customer clustering

Tin-Chih Chen*

*Corresponding author for this work

Research output: Contribution to journalArticle

1 Scopus citations


RFM model is an important method in customer clustering. Chiu and Su (2004) proposed a fuzzy RFM model to overcome the shortcomings of traditional RFM models. However, there are some problems unsolved in Chiu and Su's approach. For example, the number of customer clusters cannot be specified in advance; the inherent structure of customer data which is unknown yet valuable information to the business is not considered in forming customer clusters. To deal with these problems, a fuzzified RFM model is proposed in this study by incorporating the fuzzy c-means approach, which is based on the inherent structure of the data itself. The number of customer clusters can be arbitrarily specified in advance, considering the scarcity of marketing resources and the diversification of marketing strategies. Besides, exploring the content of each customer cluster provides the business with many meaningful suggestions that could be usefully employed to establish target marketing programmes. The example in Chiu and Su's study is adopted to demonstrate the application of the proposed methodology and to make some comparisons.

Original languageEnglish
Pages (from-to)358-373
Number of pages16
JournalInternational Journal of Technology Intelligence and Planning
Issue number4
StatePublished - 1 Dec 2012


  • Customer clustering
  • Fuzzy c-means
  • RFM model

Fingerprint Dive into the research topics of 'The RFM-FCM approach for customer clustering'. Together they form a unique fingerprint.

  • Cite this