Consumption of live streaming or entertainment content such as video games, movies, concerts, and sports events often lasts a nontrivial time period. Where are the engaging moments of contents to consumers during the period of experience? How to capture and quantify such moments? What are the business implications of such engaging moments?

Live chats become popular among leading video and live streaming platforms such as Twitch, YouTube Live, and Tencent Video. Viewers like to post real time live comments on screen during their viewership. Mining the rich live chats data, we propose a novel approach for in-consumption social listening to quantify live consumption experience. The approach is demonstrated in the context of online movie watching with a novel measure, moment-to-moment synchronicity (MTMS), to capture consumers’ real time engagement. MTMS refers to the synchronicity between temporal variations in the volume of live comments and those in movie content mined from unstructured video data such as camera motion, shot length, sound loudness, pitch, and spoken lines.

We show that MTMS strongly predicts appreciation of movies and can be evaluated at finer level to identify engaging contents. This approach can be applied to improve the timing of in-show advertising and help content producers improve the content quality.