Customer segmentation

Abstract

In general, online retailers regard the visitor traffic as a measure of success of their online shopping mall. However, traffic measures such as total number of visits per month provide little perspective regarding the revenue of online shopping mall. Since online shopping mall generally includes its own online community, it is difficult to assert that one variable effect on online shopping mall``s revenue. By knowing each factor which makes influence on the online shopping mall, the online retail shop could manage and execute marketing in a better way. This paper examines relationships between several variables of web log data and the online shopping mall``s profitability based on Internet click stream data including demographic data of customers. In this research, we identify important factors deciding profitable customer and customer heterogeneity that general online retail shops may have using customer``s web transaction data. And we classify the customers into several meaningful sub groups in marketing perspective. For this research, we perform the empirical analysis in two stages using cluster analysis: the first analysis is performed with full dataset and the second one is with member user``s data including member``s demographic data and transaction data of Korean online pet shop. Using full dataset, we identify that even the non-member users may purchase more than member user without participating in online community. Using online user``s demographic data and one``s web transaction data, we demonstrate that online customer is comprised of various kinds of purpose in visiting the site such as participating in online community or contents. As for the contribution of this paper, we try to segment customer using web transaction data rather than using user``s survey data in marketing perspective. And we suggest insights for customer segmentation in online retail shop with online community different from other kinds of retail shops.