Marketing Seminar（2016-14）

Title:Investigating the Effect of Video Advertising in Crowdfunding with Machine Learning Techniques

Speaker:Xi Li, University of Toronto

Time:Wednesday, 19 October.13:30-15:00

Location:RoomK01, Guanghua Building 2

Abstract:

This paper investigates how online video advertising affects the success rate of crowdfunding projects, using a dataset of 8,327 music projects listed on the Kickstarter website. We use machine learning techniques to measure the duration, stimulation level, and perceived credibility of the video advertising. All other things being equal, we find a lower success rate for projects with longer videos, indicating the presence of the tedium effect. Moreover, the tedium effect was stronger when the creator had prior crowdfunding experience, but weaker when the project target was higher. Second, there existed an optimal level of stimulation. The project success rate was lower when the stimulation level was either too high or too low. Finally, the project success rate increased when the video advertising demonstrated higher credibility by including humans and/or instruments in the videos. The machine learning techniques that we use prove effective in extracting useful features from a large number of advertising videos to provide new insights for the advertising theories. The results offer functional guidelines for the optimal design of advertising in the digital age.

Introduction:

Xi Li is a PhD candidate in marketing at the Rotman School of Management, University of Toronto. His research interests include big data, machine learning and marketing strategy. Recently, his research focuses on applying the state-of-the-art machine learning techniques to solve marketing problems in the digital age. Previously, he obtained a BE in computer science from Tsinghua University and an MPhil in operations research from HKUST.

Your participation is warmly welcomed!