This paper develops and estimates a discrete choice structural model of demand to address the unique challenges of the domestic film industry. The industry is marked by highly seasonal demand fluctuations and features products with clearly defined characteristics and short periods of viability. The model described herein, which is built upon the framework of BLP (1995), accounts for utility decay over time, seasonal effects, and heterogeneous consumer preferences over genre, MPAA rating, and critic reviews. I approximate the optimal instruments in a nonlinear setting, enabling the identification and precise estimation of the random coefficient parameters. These parameters shape characteristic-driven substitution patterns, a feature of the market largely ignored in past work. Using the model estimates, I identify the most over- and under-used markets in the industry by simulating the potential revenue of a hypothetical film across potential release dates. I find that the Christmas holiday season and the early portion of the year are often underutilized, while the fall markets are generally oversaturated. The paper concludes by constructing the film studios’ release date scheduling problem and solving the inherent integer-programming problem. I find a restructured release schedule that increases revenue by 17% at the firm level.