How data science can make Hollywood more diverse

Hollywood has a diversity problem, and Wharton operations, information, and decisions professor Kartik Hosanagar wants to use data science to fix it.

Digital clapperboard against a background of ones and zeros indicating big data.

Inspired by his lifelong passion for storytelling and filmmaking, Hosanagar has launched Jumpcut, a startup to help Hollywood create more inclusive content by relying on data to show industry leaders that audiences are hungry for a wider range of representation. The business also serves as an incubator for undiscovered, diverse talent.

“However you measure it, the industry has not been particularly inclusive,” Hosanagar says. “There’s a cost to audiences, which is we’re stuck in sequel culture and the lack of fresh, original stories. What we’re trying to do is really turn that on its head. We’re using data to discover new voices and stories and not just wait for Hollywood agents to discover them.”

The film and television industry has come under intense scrutiny in recent years for its lack of representation at all levels, from the actors on screen to the writers, producers, and directors behind it. McKinsey & Company released a report in March that determined Hollywood could make an additional $10 billion a year if it addressed persistent racial inequalities, and a 2020 study from Nielsen faulted streaming, broadcast, and cable television for its dearth of Black, Hispanic, Asian, and other underrepresented identities.

Hosanagar, who is also faculty director of Wharton’s AI for Business, doesn’t believe Hollywood became entrenched in its old ways out of overt bias. “The industry, as it’s become more and more of a big business, has become increasingly risk-averse, so there is a reliance on doing what has always worked in the past,” he says. “Our big a-ha moment was to recognize that there are other ways to de-risk stories and storytellers, and data is extremely good at that.”

Read more at Knowledge@Wharton.