Segmentation and Ranking of Online Reviewer Community

Abstract

Online reviewer societies flourish on contributions from different reviewers, who display a wavering engagement behavior. Effort has been made in the e-marketing literature for segmenting individuals with the help of their engagement behavior. In this study, the authors segment the reviewers of a popular travel website (TripAdvisor) through k-means clustering based on three dimensions (F-frequency, H-helpfulness, R-recency), resulting in four different reviewer segments-valuable, trustworthy, new and valueless. The authors calculate the reviewer value using fuzzy AHP and then rank the reviewer segment accordingly. The authors find that the valuable reviewers, who post eWOM regularly and get greater helpful votes by eWOM readers, are the most important. Surprisingly, the trustworthy, who also get more helpful votes with higher eWOM volume, but not posting any review recently, are the second most important. This research is a novel effort on reviewer segmentation and gives valuable insights to e-marketers.