The ubiquity of WOM in the business world underscores how instrumental it has been as a consumer engagement lever. Therefore, it is imperative for business to understand the consequential role of WOM in influencing consumer behavior. There is also a great need to improve the quality and quantity of online reviews. I address three overarching questions in my dissertation: (1) What is the effect of WOM on consumer decision making; (2) How to detect fake reviews using the review text; and (3) How to encourage reviewers to reveal their identity and give higher quality reviews.

In the first study, I estimate the causal effect of online WOM on consumer demand and uncover its mechanism in affecting the consumer decision making process. I utilize a natural experiment to examine the causal effect of online WOM on consumer demand. The setup allows me to gain granular understanding of how WOM affects the consideration set size and session duration. In addition, the availability of provider location in the dataset allows me to estimate the impact of online WOM on the consumer’s willingness to travel. In the second study, I detect fake online reviews. To identify fake reviews, I use an incidental honeypot that attracts fraudulent behavior by opening low-cost channels for fraudsters. This allows me to build a large training dataset for the machine learning classifier. Finally, in my third study, I explore how email message framing and assurance of user privacy affect the response rate and response quality of online WOM. I conduct a field experiment to uncover how the propensity of a user to give feedback for their doctor can be influenced by a motivational message and how privacy assurance affects identity revelation. Collectively, these studies advance our knowledge on the antecedents and consequences of online reviews, which helps business and society to better utilize WOM for greater value creation for consumers.