As a quantitative marketing researcher, my primary research interests lie in online retailing, online word-of-mouth, user-generated content, and AI technology.
Methodologically, my research leverages methods of causal inference, machine learning, and structural modeling.
My research agenda :
(1) Exploring the impact of spatial proximity among consumers in offline settings on their online purchasing behaviors.
(2) Investigating the impact of generative AI disclosure policies on user engagement and influencer strategies.
(3) Examining how online reviews and critics, distinguished by demographic factors such as gender and ethnicity, influence product demand dynamics over time.
(4) Analyzing how live comments affect the design and performance of entertainment products.
In this paper, we investigate how spatial proximity can affect consumers’ purchase behaviors and how an e-commerce platform can leverage this effect to increase sales. Spatial proximity, defined as the geographic closeness of consumers, plays a significant role in their purchase behavior. However, existing literature mainly focuses on its effect on new product adoption while largely ignoring other consumer decisions, such as purchase quantity and inter-purchase time, which are equally critical for online retail operations. Further, the impact of consumer spatial proximity on supply-side marketing strategies, as well as its dynamics and heterogeneous impact, remains understudied. Based on social identity theory, this study examines these issues using a unique dataset from a leading e-commerce platform in Hong Kong. The results indicate that consumer-to-consumer (C-C) spatial proximity has a significant impact on consumer online purchases (when and how much to buy) and the platform’s marketing strategies (price and promotion), and that such effects can change over time. Further, significant heterogeneity exists across consumers and product categories with different levels of social identity. The findings suggest that the focal platform can increase its annual revenue by more than $5,218,541 (an increase of 6.37%) by using C-C spatial proximity-based price and promotion strategies for one category (personal care and health) alone. Consequently, this study provides important theoretical and managerial implications for marketing researchers and practitioners in online retail operations.
This study employs a Multivariate Dynamic Linear Modeling (DLM) approach to dissect how online reviews and critics distinguished by demographic factors like gender and ethnicity influence product demand dynamics over time. Analyzing data from two distinct categories—movies (hedonic) and cameras (utilitarian)—this research unveils the intricate roles of message source expertise and demographics in shaping consumer demand. There are several key findings. First, we find that both review and critic valences significantly boost demand, albeit with nuanced, time-sensitive effects that vary by product type. Critic valence exhibits enduring positive influence, contrasting with the transient impact of review valence in the movie sector. Second, our findings challenge conventional gender stereotypes; showing that the percentage of women reviewers and women critics have a positive impact on the product demand and the effect gets stronger over time for reviews written by women. However, we identify a concerning trend: reviews from minority groups depress sales figures, a bias somewhat alleviated when minority voices belong to expert critics. This points to the mitigating power of perceived expertise against racial biases since it provides a clue of competence. In addition, the interaction effect between reviewer demographics and review valence shows that the percentage of posts by women and minority group would strengthen the impact of valence on product sales except for the nonwhite reviewers. The findings related to the demographics stay consistent for both movie and camera category. Finally, our research provides important managerial and practical implications when it comes to recognizing the most influential strata of the reviewers or critics.
As a disruptive innovation, the emergence of generative AI has dramatically changed the way users generate their content, which can have a significant impact on users, audiences as well as the general environment of social media platforms. However, existing literature has not thoroughly investigated the impact of GenAI disclosure on social media platforms. In this paper, we explore the impact of generative AI (GenAI) disclosure policy. In our research, we leverage TikTok’s official announcement of a function of labeling AI-generated content on the platform and use a difference-in-difference (DID) approach to examine the impact of this policy. More specifically, we investigate 1) what’s the impact of the AI policy on supply-side influencer strategy of whether to apply GenAI in their posts, the content quantity, the content quality as well as the content variety; 2) what’s the impact of the AI policy on demand side consumer engagement in terms of volume of comments and valence of comments; 3) what are the dynamic impact of the policy over time; 4) how would influencer status and content category moderate this impact.