Using Data Mining Techniques to Discover Customer Behavioural Patterns for Direct Marketing

Mobile telecommunication services are becoming increasingly popular because of rapid technological advancements, especially in the areas of service development and fulfilment of customer requirements. Undoubtedly, mobile telecommunications have become a profitable but fiercely competitive industry. The key to survival in this competitive market lies in knowing customers better and retaining every valuable customer. Due to the advancement in database technology and the availability of huge amounts of customer data, marketers have started to utilize powerful data mining tools in direct marketing to delve deeper into their customers’ nature and habits. This strategy enables them to personalize marketing and promotional information which they can then send to their customers. There are mainly two types of data mining techniques used in practice: supervised learning and unsupervised learning. In this study, the main research objective is focused on the use of unsupervised learning techniques to support personalized direct marketing activities in the mobile telecommunication industry. The study includes empirical studies using both existing technologies and theoretical development of novel methodologies in the field of data mining and marketing. Our analysis was important to the marketers of mobile telecommunications. Our approaches can be adopted in actual practice to discover behavioural patterns of customers so that services can be designed to better fit their needs. Marketers can also monitor customers’ behavioural changes so that they can make strategic decisions beforehand and, subsequently, retain valuable customers. The methods adopted and proposed in our research may inspire new studies on issues such as how to evaluate and interpret clustering results, how to identify changes in customers’ behavioural patterns using clustering techniques, and how to combine clustering and classification tools more efficiently.