Document Type : Original Article

Authors

• Masoumeh Hoseinzadeh Shahri ¹
• Fatemeh Ghobaee arani ²

¹ Associate Professor

² Master's degree in Management Department, Faculty of Social and Economic Sciences, Al-Zahra University, Tehran, Iran

Abstract

Consumer behavior is a critical aspect of marketing strategy that involves understanding customers' buying habits, motivations, and preferences. Better understanding of customer behavior through innovative methods of storing and analyzing customer data and information leads to formulation of more effective strategies. The emergence of new computing technologies has brought about major changes in the ability of organizations to collect, store and analyze big data. Many researches have done customer classification using unsupervised machine learning algorithms such as K-Means using the famous RFM model, but these models are insufficient by ignoring other important parameters according to the field of application. This research is applied in terms of purpose and descriptive in terms of data collection, and it is a quantitative type of research that was conducted using 200,000 transactions of online retail store customers during the period of 2013 to 2018. The model is modified by adding variety "D" as a fourth parameter, referring to the variety of products purchased by a given customer. Classification based on RFM-D is applied in the online retail market in order to identify behavioral patterns for customers. Examining the behavior of cluster customers showed that the variety of products, along with other behavioral variables, provided more profitability than the RFM model.

Keywords

• Customer buying behavior
• RFM-D
• diversity
• K-Means algorithm
• customer classification