Personality-Based Content Engineering for Rich Digital Media
With increasing consumer resistance to advertising, firms have turned to rich digital media, such as videos and photos, to attract attention and boost brand awareness. Although extant research may help firms promote these media more effectively once they are released, the reality of the marketing process is that it begins with the creation of the media. Thus, firms may benefit from understanding what media is likely to achieve greater popularity, based on its content features. We develop an approach to understand the role of content on the consumption of online videos, using a unique dataset including 16,414 videos from 363 YouTube channels. We introduce an algorithm to identify content features of online videos that predict high performance, and apply this algorithm both globally and on a channel-level. Our approach categorizes videos by their performance relative to comparable videos, and leverages random forests to identify content features associated with level of performance. We test this approach through an analysis of the personality of online videos, using NLP to assess the extent to which videos exhibit each of the “big five” personality traits. Analysis reveals that videos associated with high-performing personalities generate a statistically significant increase in views, suggesting that our methodology can offer prescriptive insights that inform content engineering.