This study analyzes the purchase patterns of MPEG Audio Layer-3 Players (MP3) customers in Korea. Factor analysis, clustering, and association rules are used to find the purchase patterns of segmented groups. From our analysis, 7 major factors were identified: technique, business, stylishness, rationale, late adaptor, pragmatic propensity, and music. Based on these factors and demographic data, six types of user groups are identified. Subsequently, we identify the chain of purchase processes of each cluster of customers. Finally, we propose target marketing strategies for segmented groups. We expect that our study results can provide direction for marketing and public relations strategies for MP3P manufacturers.
MPEG Audio Layer-3 Players (MP3Ps) have become a cultural icon for youth and their popularity has spread all over the world based on attractive designs and various functions such as wireless Local Area Network [Wi-Fi] (LAN), Digital Multimedia Broadcasting (DMB), video players, etc.
According to IDC research, the worldwide MP3P market was no more than 26,000 in 2001, but it reached 1,500,000 by an annual growth rate of 176% until 2005. With the recent extension of the MP3P market – dominated by the leading company, Apple – other large global enterprises such as Microsoft, Samsung, Nokia and Matsushita are advancing or announcing plans to pursue a greater proportion of the MP3P market. Therefore, distinguishable marketing strategies are required to successfully take and/or hold a dominant position while participating in a fiercely competitive global market.
Many studies have used data mining techniques to increase the efficiency of marketing. Sohn and Kim (2008) identified patterns of use for additional services of mobile telecommunication subscribers in Korea by using quantitative association rules. Chang, Hung, and Ho (2007) proposed a forecasting model for potential customers’ purchasing behavior based on clustering and association rules. Huang, Chuang, Ke, and Sandnes (2008) used back-propagation neural network with association rules to predict what customers will buy next and from what categories. In multiple relations, multi-level association rule analysis is encouraged. Therefore, multi-level association rule analysis is suitable for our paper, which includes various information resources, such as purchase locations, prices, and satisfaction levels.
The main purpose of this study is to identify characteristics of target customers based on lifestyle variables (Lawrence & Giles, 2000), which include usage and various opinions about MP3Ps. We use factor analysis for clustering customers. Next, we employ association rules to understand preferences and purchase patterns of each cluster. Finally, we propose target marketing strategies for segmented groups.
The construction of this paper is arranged as follows: Section 2 describes the proposed research methodology. In Section 3, we conduct the empirical analysis using data mining methods and present marketing strategy for each cluster. Finally, Section 4 summarizes the study results and describes the potential for further study.
Development of computer technology and communications make possible the storage and analysis of huge amounts of customer data. By analyzing such a large quantity of data, we can discover each individual customer’s needs and interests. Thus, it could be possible to build a marketing strategy for potential customers. In other words, an enterprise with a massive database of customer information to analyze could supply services according to customers’ demands. Crucial managerial tasks for the enterprise would include categorization of target customers into different segments based on their individual characteristics and purchase patterns, and analysis of these different segments.
We conducted factor analysis and cluster analysis to identify characteristics of target customers based on lifestyle variables which include usage and various opinions about MP3Ps. We also conducted cluster analysis to define characteristics of target customers based on factor and demographic data. Finally, we identified the relationship between purchase patterns and the processes used by customers to obtain information about MP3Ps, using association rule analysis in order to propose marketing strategies for each cluster based on the results.
Result showed that many customers have acquired information from the internet and from internet malls. These customers consider function and performance to be the most important factors. Also, they make their purchases through internet malls. Based on these result we propose a marketing strategy.
We expect that our study results can provide direction for marketing and public relations strategies of MP3P manufacturers. Also, this research offers insightful information for creating successful marketing strategies for MP3Ps based on purchase patterns.