Customer segmentation with purchase channels and media touchpoints using single source panel data

Abstract

This study examines how customers use multiple channels and media in modern retail environments. It segments customers by using Latent-Class Cluster Analysis, which focuses on the purchase channels of bricks-and-mortar and online stores, media touchpoints of PC, mobile, and social media, and psychographic and demographic characteristics. It extends the framework of prior research by analyzing 2595 Japanese single source panelists’ data in which purchase scan panel data on low-involvement, more frequently purchased categories, media contact log data, and survey data are tied to the same ID. The analyses reveal seven segments including the properties of research shoppers and multichannel enthusiasts.