Consumer Segmentation Analysis of Multichannel and Multistage Consumption: a Latent Class Mnl Approach

The prevalent multichannel shopping environment is driving many consumers to choose between online and offline channel at the information search and product purchase stages of a shopping experience. Consequently, multichannel retailers face the challenges of identifying different target consumer groups and maximizing the value of each channel by understanding and serving each group more effectively. Hence, this study attempts to identify consumer segments by examining consumers’ perceived channel values at different shopping stages. The latent class MNL (LC-MNL) method, as a powerful tool that is able to detect consumer heterogeneity in the same consumption scenarios, is applied to conduct consumer segmentation analysis based on the consumer’s perceived values, including channel benefits and costs, as well as different channels’ characteristics. By using the survey data of 1325 consumers, results indicate two segments comprising innovative consumers and conventional consumers in terms of online vs. offline channel usage. Furthermore, the logit regressions for segments estimation illustrate that the two segments are significantly different in terms of channel attributes and consumers’ intrinsic channel preferences. This study contributes to the extant electronic commerce and multichannel marketing literature by designing a rigorous consumer segmentation method which incorporates both interpretation and prediction capabilities and analyzing the underlying influential factors for different segments. The results can further provide useful guidance to marketing and sales practitioners in designing effective channel attributes to meet the needs of consumers belonging to different segments.