Research

Abstract: Thumbnails serve as representations of video content when viewers are "in the dark" before watching. Yet it remains unclear whether thumbnails should summarize the underlying video, like a synopsis, or tease the video. We study how thumbnails, relative to the video content they represent, affect video decisions. We propose a thumbnail-video mining procedure that uses multimodal LLMs and computer vision to transform unstructured thumbnails and video content into interpretable features, and then construct theory-based measures to characterize both the thumbnail itself and its relationship with the video it represents. We use secondary data from YouTube to document real-world relationships between thumbnails and video performance, and then experimental data from CTube, a video platform we built to randomize thumbnail exposure, to estimate a joint model of video choice and watchtime. Our results reveal a fundamental click--watchtime tradeoff. ``Teaser-style" thumbnails that are visually appealing and emotionally engaging increase clicks, but these features do not sustain viewing. Instead, thumbnails affect watchtime through how the video unfolds relative to the thumbnail: viewers are more likely to quit when video content diverges from the thumbnail, but this penalty weakens after viewers observe the moment revealed by the thumbnail in the video. We simulate performance of alternative thumbnails and show that there is no one-size-fits-all thumbnail. Effective thumbnails require balancing visual impact with content alignment and timing, while tailoring each thumbnail to viewers and video content.

Abstract: How effective is platform self-regulation at eliminating harmful content? We examine Twitch’s ban on unlicensed online gambling livestreams implemented in October 2022. Using a novel panel dataset covering the top 6,000 Twitch streamers, we identify banned content and affected streamers by leveraging video analysis of historical clips, high-frequency stream titles, and in-stream chats. On the supply side, the policy led to a 63.2% reduction in weekly gambling streams and a 44.3% decrease in overall streams among streamers whose content was banned. Streamers whose gambling content was not banned also reduced their gambling and overall content, indicating spillover effects from the policy. This reaction was more pronounced among popular streamers and those with greater reputation concerns. On the demand side, while the policy reduced total viewership, low-tier subscriptions, and engagement, revenue from high-tier subscriptions (reflecting more loyal viewers) remained unaffected. The policy also reduced the gambling viewership for viewers classified as minors, although the magnitude was slightly smaller than for other viewers. Our findings highlight both the potential and limitations of platform self-regulation: although targeted bans can substantially reduce targeted content and its engagement, complementary actions may be necessary to mitigate unintended side effects and ensure effective protection for specific user groups.

Abstract: We study content collaboration in the creator economy, in which competing creators mutually agree to collaborate on joint content and negotiate on content production and revenue sharing. Using a game theory model with creators competing for consumers on a Hotelling line, we show that collaboration allows creators to use the jointly-produced content to moderate competition, while using their individual content to expand into new audiences. This increases content diversity but also leads to increased monetizability of content. In general, collaboration among creators has an effect of increasing the profits of creators while reducing consumer surplus. When creators create content with heterogeneous entertainment values, the creator producing content of lower entertainment value has an incentive to free ride on the collaborative content. This free riding may increase surplus for consumers (who without collaboration would watch content of low entertainment value), thereby improving creators' profits as well as consumer surplus. Our results provide guidance to content creators, to platforms designing tools to facilitate collaborations, and to policy makers.

Abstract: We build a Bayesian learning model to investigate how thumbnails, previews of video content, affect consumers’ reactions to videos in terms of video choice and watchtime via two roles: thumbnails as expectation-based reference points that shape viewers’ expectations of the video content before they start watching a video, and thumbnails as informational reference points that build viewers’ anticipation for a video. We model consumers' decisions to click on a video and continue watching the video as based on their priors (the thumbnail) and updated beliefs of the video content (the video's frames, characterized as multi-dimensional and correlated video topic proportions). We create a novel video streaming platform called "CTube" (a simplified version of YouTube) to conduct a novel field-framed online experiment by randomizing thumbnails viewers see to start a video. Leveraging the high-frequency clickstream data tracked by our video platform, we estimate the bayesian learning model by restoring from clickstream data individuals’ exposed video thumbnails, video choice and watchtime decisions. Our results suggest that  viewers overall prefer watching videos longer when there is a higher disconfirmation between their initial content beliefs based on the thumbnail and updated beliefs based on the observed video scenes (signals). In addition, viewers prefer less content disconfirmation before observing the thumbnail, highlighting that the role of disconfirmation may change before and after viewers observe the moment highlighted by the thumbnail. Based on the model's estimates, we then run a series of counterfactual analyses to propose optimal thumbnails and compare them with current practices of thumbnail recommendation to guide creators and platforms in thumbnail selection.