Customer portfolio analysis using the SOM

Abstract

In order to compete for profitable customers, companies are looking to add value using Customer Relationship Management (CRM). One subset of CRM is customer segmentation, which is the process of dividing customers into groups based upon common features or needs. Segmentation methods can be used for customer portfolio analysis (CPA), the process of analyzing the profitability of customers. This study was made for a case organization, who wanted to identify their profitable and unprofitable customers, in order to gain knowledge on how to develop their marketing strategies.Data about the customers were gathered from the case organization's own database. The Self-Organizing Map (SOM) was used to divide the customers into segments, which were then analyzed in light of product sales information.

Combining visual customer segmentation and response modeling

“…However, common motivations for using the SOM over alternative methods are the interaction between the two tasks of clustering and projection, the predefined grid structure for linking visualizations, flexibility for missing data and computational efficiency (see e.g., Vesanto, 1999). In addition, the SOM has previously been shown to be an efficient and easy-to-interpret tool for customer segmentation (Kim, Wei and Ruys, 2003;Holmbom, Eklund and Back, 2011;Yao, Holmbom, Eklund and Back, 2010;Vellido, Lisboa and Meehan, 1999;Lee, Suh, Kim and Lee, 2004;Lee, Xiang and Jing, 2005;Lingras, Hogo, Snorek and West, 2005).…”

“…Effective segmentation enables companies to interact with customers in each segment collectively, and allocate limited resources to various customer segments according to corporate strategies. A range of data mining techniques have been used for customer segmentation, e.g., decision trees (Kim, Wei and Ruys, 2003), self-organizing maps (SOMs) (Holmbom, Eklund and Back, 2011;Mo, Kiang, Zhou and Li, 2010;Yao, Holmbom, Eklund and Back, 2010), k-means clustering (Dennis, Marsland and Cockett, 2001;Hosseini, Maleki and Gholamian, 2010), and combinations of different methods (Kuo, Ho and Hu, 2002;McCarty and Hastak, 2007). These studies have successfully demonstrated the usefulness of customer segmentation in a variety of industries.…”

Direct mailing decisions based on the worst and best practice cross-efficiency evaluations

“…For customers' credit (CC) rating prediction, Huanga et al (2004) applied a learning method based on statistical learning theory, support vector machines (SVM) together with back propagation neural networks (BNN). Holmbom et al (2011) used self-organising map (SOM) and identified profitable and unprofitable customers. The gained knowledge from this process is used to develop marketing strategies.…”

Rating of Enterprises’ Activities by the Modified Cluster Method

“…The usage of statistical methods for data analysis requires an easy visualization of research results, which can be achieved by using the Kohonen neural networks (Kohonen's self-organizing maps -SoM) (Debok, Kohonen, 2001;Kohonen, 2008). Methods of self-organizing maps are widely used in economic research, in particular, for the analysis of consumer portfolio (Holmbom et al, 2008), of the marketing technology (Nikishina, 2003). The cluster approach was used to delimite the regions by the human capital requirements (Bezrukov, Kolosova, 2008) while studying the financial and economic situation of construction enterprises (Kovalenko et al, 2010), etc.…”