讲座题目：Optimal Freemium Pricing of Digital Content via Large-scale Randomized Field Experiment and Content Analysis
主 讲 人：Natasha Zhang Foutz
Natasha Zhang Foutz博士是美国佛吉尼亚大学麦金泰尔2003网站太阳集团的终身教职副教授，她的研究聚焦于娱乐营销、数字媒体及移动营销等领域，主要采用机器学习、统计计量以及实地实验等方法，研究成果发表于Journal of Marketing Research, Marketing Science等顶级期刊。她目前担任Journal of the Academy of Marketing Science的领域主编,讲授营销分析、娱乐营销、营销模型等课程，曾获多个权威学术会议的最佳论文奖、Mallen Award(终身出版电影产业研究学术贡献奖)、Management Science年度最佳评审专家、佛吉尼亚大学教学优秀奖等荣誉。
The $300+ billion digital content industries, encompassing e-books, movies, music, gaming, and many others, commonly leverage freemium pricing, offering initial contents, such as the first few chapters of a book or first few minutes of a song or video for free, hoping to monetize the later content with a premium. Determining the optimal amount of free content, or conversely the optimal charging point (such as charging at Chapter 40), is challenging, since offering scant free content risks customer churn, whereas offering a lavish amount diminishes profit. More importantly, freemium pricing for digital content, unlike for consumer packaged goods (sampling literature) or technology products (versioning literature), is closely related to content dynamics. For instance, charging at a story’s emotional high or critical turning point would potentially lead to stronger customer willingness-to-pay.
Collaborating with a leading e-book platform, we determine the optimal charging points for e-books via a large-scale randomized field experiment involving 1.3 million customers. The resulting remarkable 50% lift in revenues per product from optimizing the charging point would translate into a $100+ million annual revenue gain for the entire platform. We further propose innovative text analytic methods to determine the landmarks of content dynamics of each product, such as the sentiment or emotion culminations. The model-free evidence and statistical analyses reveal that the identified optimal charging points tend to occur right after the second culmination. Two mechanisms underlie this finding: the moment-to-moment synchrony between a book’s content and consumers’ comments, as well as bulk (multi-chapter) instead of single-chapter purchase, commonly peak right after the second culmination. Finally, a personalized pricing design showcases the vast commercial power of leveraging the AI-Machine Learning driven optimization and personalization to automate the conventionally intuition-laden and labor-intensive pricing for massive numbers of heterogeneous digital content products and customers.