The dynamics of online WOM content and product sales for movies

Abstract

User generated product reviews serve as a valuable source of information for customer’s decision making. Most of earlier work studied the impact of product review on sale by measuring the volume and numeric rating of reviews. In this paper we analyzed the information embedded in product reviews. We argue that Product reviews are multifaceted, and hence the textual content of product reviews is an important determinant of consumer’s choices over and above the valence and volume of reviews. To determine this, we used Natural Language Processing (NLP) for text mining and breaking down textual reviews into categories describing different product features. We apply our approach to the movie industry because of the unique nature of movies and high economic importance of the industry. We used consumer reviews data from Internet movie database (IMDB) and examine the impact of movies features in textual data on movies sales. Our results suggest that content about lead actors and supporting actors or overall about acting in reviews before release have significant impact on sales. Also, these features have lag effect on movies revenue. However after Experience the story plot/screen play and Visual effect impact the movies sales. We also found that across genre different movies features have significant relationship with movies revenue. This paper demonstrates how textual data can be used to learn how different product feature impact sale