Doctor of Philosophy (PhD)



Document Type



Technology development and e-commerce growth boost the impact of online reviews. Online reviews have become an important element of the marketing communication mix. To design effective marketing campaigns for better marketing outcomes, managers must understand how to interact with customers via online reviews and how customers’ behaviors are affected by online reviews. In this dissertation, I explore two questions in this domain: companies’ online interaction with customers through online company responses to customer reviews, and customers’ new product adoption process influenced by online reviews.

Essay 1 examines why companies respond to reviews, and whether and how online company responses improve review quality on online platforms over time. Applying text-mining techniques combined with a big data analysis, we measure the review quality with two dimensions: review comprehensiveness and review readability, examine the textual characteristics of reviews that drive companies to respond to reviews, and explore how behavior and content cues of online company responses impact review quality over time. Leveraging a big data set from an e-commerce platform, we find that: 1) companies are more likely to respond to reviews with lower sentiment and higher sentiment deviation scores; and 2) online company responses improve review comprehensiveness over time but do not increase review readability. We further verify that longer responses enhance the improvement of review comprehensiveness while more tailoring responses amplify the increases in review readability, and response intensity doesn’t improve review quality. Our findings could guide companies to improve their online communication strategy and enhance online customer engagement quality.

Essay 2 empirically investigates the differential effects of online review characteristics on new product adoption for two types of consumers (innovators and imitators) and how such effects may be altered by different dimensions of review characteristics. Using a data set collected from Amazon, we find that the differential effects exist, but are moderated by the dimensions of review characteristics: the numerical review characteristics (i.e., review volume, review rating, and rating variation) have larger effects on imitators than on innovators, but the review textual characteristics (i.e., review emotional content) show the opposite effects. We further discuss the managerial implications of our findings.



Committee Chair

Wu, Jianan



Available for download on Monday, May 21, 2029

Included in

Marketing Commons