Funded PhD Research

Dokyun Lee

Effective Marketing Content Strategy on Facebook: Industry and Product Level Investigation

Operations and Information Management Department; Faculty Adviser: Kartik Hosanagar

This research builds on prior research funded by the Baker Retailing Center: The Effect of Advertising Content on Consumer Engagement: Evidence from Facebook. The first paper investigated what content (e.g., price, deal, emotional, philanthropic) in companies’ posts on their Facebook pages drives more engagement, as measured by the number of likes and comments, with post-day-level panel data spanning more than 100,000 unique messages across about 800 companies over an eleven-month period. The research employs cutting-edge natural language processing in combination with econometric methodology to analyze large-scale text data with causal implications.

This paper extends previous research to study how variety of different product attributes as well as industry-level attributes moderate the effect of content strategy for engagement. While the first paper examined the content strategy at the higher level, this paper investigates this at a more detailed level, providing more direct strategy suggestions for retailers based on what industry segment they are in and the kinds of products they want to market.

2. Comparative Study of Different Recommender Systems on Sales Volume and Diversity
Operations and Information Management; Faculty Adviser: Kartik Hosanagar

While many types of recommender systems such as collaborative filtering (e.g.,’s “customers who purchased this item also purchased”), content-based, and bestseller-based systems are being utilized in e-commerce, both academics and practitioners don’t know their comparative impact on sales volume and sales diversity. Sales volume refers to the number of products purchased or individual consumers’ wallet sizes while sales diversity refers to the market share distribution or concentration of purchased products (e.g., niche item vs broad-appeal item). With a field experiment on a top retailer website, this research investigates comparative impact of different recommender algorithms on sales volume and diversity.

Publications, Presentations & Awards