New product sales forecasting is a crucial task in many innovative companies. Conventionally, the accuracy of sales forecasting is conditional on long-term, sufficient data of sales history. However, sales information on newly launched products is considerably limited, unavailable, or inaccessible.
Because virtually all types of product are seasonal, the main objective of this study is to propose a seasonal sales forecasting model in the settings of the U.S. motion picture market by means of online open access data, such as critiques, comments, and ratings, which express the original interests of different stakeholders of movie industry and are available at both pre- and post-released phases. Also the performance of marketing campaign via online visual advertisement such as posters or banners will be examined in terms of their visual effects of design elements.