A new approach to segmenting multichannel shoppers in Korea and the U.S.

Abstract

Advancements in digital technology and devices enlarge dimensions of e-commerce, reforming the ways that consumers shop and purchase products and services. In particular, the mixed use of online, mobile, and offline channels and devices for shopping provides B2C firms with unprecedented challenges and opportunities to develop effective segmentation approaches that capture multitude of newly emerging consumers' shopping patterns. This paper aims to classify consumers along with their shopping patterns and channel preferences by using rank order survey data from Korean and American consumers on their path-to-purchase behaviors. Cluster analysis and Association Rule Mining (ARM) are applied for segmentation and its characterization. Relative importance of path-to-purchase factors such as information search location, payment method, delivery option, and payment location are assessed to determine the differences in Korean and American consumers regarding their shopping patterns and preferences. Network visualization of rules shows the differences in shopping preference and patterns of Korean and US consumers both at micro and macro levels.