Binge Yourself Out: The Impact of Binge Watching on Viewership and Subscription of Media Products, Evidence from a Randomized Field Experiment

Binge watching is becoming a popular way to consume video. Content providers and distributers are increasingly releasing all episodes of the same season of a TV show at once, allowing consumers to watch them in one sitting. This strategy is likely to change the way the media industry operates but little is still known about how binge watching affects consumer behavior and satisfaction. In this paper, we partner with a major telecommunications provider that offers Video-on-Demand (VoD) to study how consumers use binge, how much they enjoy VoD that allows for binge watching and whether binge watching affects their willingness to pay for VoD. We use data from two surveys and from two household level randomized field experiments. The first survey shows that consumers are willing to pay a significant premium for a service that allows them to binge watch their preferred TV show. As a follow up from this survey, in our first experiment, a random set of households were given access for free to a TV channel that broadcasts TV shows 24/7. A random subset of these households could consume this channel using time-shift TV and thus binge watch past episodes. Therefore, this service resembled subscription-based VoD with a small catalog. This experiment shows that consumers that binge watch consume more of this TV channel but subscribe it less after the gift period. We hypothesize that catalog exhaustion could be a driver for the observed effect. To further test this hypothesis, in our second experiment, a random set of households were given access for free to a subscription-based VoD catalog as large as Netflix's in Europe. We find that households that binge watch from this catalog are also less likely to purchase the service after the gift period and this result is smaller in magnitude than in the previous case. We also show that these consumers enjoyed their experience with the VoD system more than consumers without access to this gift, thus ruling out the concern that the content offered as part of this catalog could be of no interest to households in our sample. Finally, the second survey shows that lack of content refresh is one of the main reasons why these consumers do not want to pay to access this service and this concern is more prevalent among consumers who binge watched during our second experiment. In sum, we find that binge watching might yield surprising effects that content providers and distributors need to take into account when designing their strategies to release content. In particular, they need to consider whether consumers that binge watch exhaust VoD catalogs more quickly. This would require content providers to turnover content much faster, which could call for significant changes to the supply side of the market.