RESEARCH PROJECTS

 

with Puneet Manchanda and Eric Schwartz

A majority of US households view on-demand content on streaming video services. Not surprisingly, ad spending on these online services is growing rapidly. However, extant research on streaming media has not explored the balance between the interest of the viewer (content consumption) with that of the platform (ad exposure). We do this using two new metrics that capture the interplay between content consumption and ad exposure using viewing data on a streaming media platform. The first metric (Bingeability) measures non-linear content consumption while the second metric (Ad Tolerance) measures the willingness of a viewer to continue viewing after ad exposure. Using causal machine learning methods that comprise a tree-based model with instrumental variables, we capture the impact of ad delivery on Bingeability and Ad Tolerance for individual viewers for each viewing session. The results indicate that the “sweet spot” that balances the interest of the viewer and the platform consists of short (ad) pod durations that are equally spaced at longer intervals during a viewing session. We discuss the implications of our results for managers of streaming platforms.

Keywords: Advertising, Streaming Media, Binge-Watching, Machine Learning, Causal Inference

VIDEO INFLUENCERS: UNBOXING THE MYSTIQUE

with Puneet Manchanda 


Influencer marketing is being used increasingly as a tool to reach customers because of the growing popularity of social media stars who primarily reach their audience(s) via custom videos. Despite the rapid growth in influencer marketing, there has been little research on the design and effectiveness of influencer videos. Using publicly available data on YouTube influencer videos, we implement novel interpretable deep learning architectures, supported by transfer learning, to identify significant relationships between advertising content in videos (across text, audio, and images) and video views, interaction rates and sentiment. By avoiding ex-ante feature engineering and instead using ex-post interpretation, our approach avoids making a trade-off between interpretability and predictive ability. We filter out relationships that are affected by confounding factors unassociated with an increase in attention to video elements, thus facilitating the generation of plausible causal relationships between video elements and marketing outcomes which can be tested in the field. A key finding is that brand mentions in the first 30 seconds of a video are on average associated with a significant increase in attention to the brand but a significant decrease in sentiment expressed towards the video. We illustrate the learnings from our approach for both influencers and brands.

Keywords: Influencer Marketing, Brand Advertising, Social Media, Interpretable Machine Learning, Deep Learning, Transfer Learning