Online review is an important form of electronic word of mouth (eWOM) that helps customers make purchasing decisions. In a set of reviews, the review with the most helpfulness votes are seen as most helpful. While researchers have demonstrated how review and reviewer characteristics influence helpfulness votes, a largely uninvestigated issue is how herding behaviors can influence customers’ voting participation and direction. Drawing on herd behavior literature, we propose that review voters will discount their own information when faced with clear and strong signals from previous voters. Thus, they will herd previous voters’ voting direction. On the other hand, review voters will value their own judgments when faced with weak signals from previous voters. Herding can influence both a voter’s perception of a review’s helpfulness and his/her vote. This research extends review helpfulness literature that herd behaviors could moderates customers’ perception of review helpfulness and voting direction.