Online word-of-mouth information markets hosted by retailers like Amazon have matured and become well-accepted. A salient concern however is whether or not the current setup for delivering user reviews is efficient in helping prospective consumers find their product fit. A typical customer relies on the retailer to sift through the sheer amount of reviews and avoid information overload. In this study, we investigate the role of review featuring by the retailer and helpful voting by the crowd in facilitating efficient information consumption. We show that these tools are critical determinants in the success of this market. We find that the greater the divergence between the signals from crowd endorsed featured reviews and the crude review average, the greater is the deviation in consecutive reviews from the general review valence. Thus, this study serves as the first step in understanding the importance of such featured reviews which we label as "superstar" reviews.