The Dynamics of the Buzz: How do online reviews change over time?

• Main Authors: KIM, Youngsoo, Hu, Nan, KOH, Noi Sian
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2010
• Subjects: Computer Sciences; E-Commerce
• Online Access: https://ink.library.smu.edu.sg/sis_research/1616
• Institution: Singapore Management University

Online review sites contain vast amount of information for knowledge discovery of consumers' attitudes and preferences. This is especially so for popular products where the number of reviews can be in thousands. In presence of strong network effects, it is crucial to understand the review patterns and sentiments of customers so as to devise effective business strategies. Thus, this paper addresses the following questions: 1) what are the underlying patterns of online reviews over time? 2) How does it differ for products of different popularity? and 3) How does it differ for products of different categories? As most of the existing research has focused on the quantitative aspects of online reviews, such as ratings, we fill in the gap by analyzing the content of reviews. In particular, we examine the dynamics of review for products of different popularity/category based on the reviews' quantitative and qualitative characteristics. Our approach explores multiple features of review content - writing style, sentiments, readability and persuasiveness. Our results show that the quantitative and qualitative features of the reviews for books that are most popular (or factual) do not have large variations over time. On the other hand, the reviews for books that are least popular (or fictional) have larger variations over time. Hence, if subsequent reviews for products are merely restating the early reviews, the usefulness and impact of the subsequent reviews would be marginal as compared to the early reviews. However, when subsequent reviews do have different attributes from the early reviews, firms have to treat reviews differently and construct different strategies depending on the life cycle of the product. An understanding of the review dynamics can help firms to better predict the impact of reviews and utilize such impact to boost the product sales.