Payal Prem Kumar examined the relationship between Tesla’s stock prices and CEO Elon Musk’s connection to President Trump
Payal Prem Kumar
It’s a foundational debate in boardrooms across America: Does a controversial CEO help or hinder a company’s bottom line?
By taking a test case of Elon Musk and Tesla, Daniels College of Business student Payal Prem Kumar turned this existential question into objective data—with help from artificial intelligence.
“Mixing marketing with machine learning is very important. Because if you know what you’re doing, if you can prompt your AI correctly, you can increase your reach by a crazy amount,” said Prem Kumar, who graduated in June.
Fresh off completing a dual master’s degree in business analytics and marketing, Prem Kumar developed her analysis as part of an independent study project at the Consumer Insights and Business Innovation Center (CiBiC), Daniels’ own on-campus consulting firm. She drew her inspiration from her undergraduate years, during which she originally pursued a mechanical engineering degree at the Rochester Institute of Technology’s Dubai campus, before making the switch to business.
“I really do enjoy playing with artificial intelligence and machines, coding and identifying what’s going on,” Prem Kumar said. At a time when many jobs are being swallowed up by AI-driven efficiencies, she decided to learn how to drive the machine.
As an international student from Dubai, Prem Kumar didn’t have friends or family nearby when she first arrived two years ago, but she fell in love with Denver’s abundant sunshine and access to the outdoors. She quickly felt at ease making friends and developing mentorships on campus, where she served as president of the DU Marketing Club.
“DU has given me so much more than I could ever comprehend,” Kumar said. “I like to say that I’ve built a family here.”
Prem Kumar’s interest in Tesla stemmed from her experience making sense of American politics and culture—from Musk’s presence in the news cycle and his connection to President Donald Trump, to how it might relate to the performance of Telsa’s brand. To pursue her project, Prem Kumar worked closely with Melissa Archpru Akaka, co-director of CiBiC and associate dean for research and brand strategy.
“The intersection between marketing and analytics is a really open playing field,” Akaka said. “The more we can think about consumer behavior in an analytical way and look for different factors that might impact brands, there’s a lot more discovery to be made.”
Prem Kumar used the coding language Python to analyze public comments about Musk and Tesla, scraped from the popular social media platform, Reddit.
The easy anonymity of users and lack of restrictions for data scraping, combined with the site’s narrowly focused community pages, allowed Prem Kumar to analyze approximately 780 user posts. She organized the data according to key words and dates, drawing valuable insights into Tesla’s performance over time.
“Reddit was perfect. Everyone is so open, people just type what they want to,” Prem Kumar said.
The posts provided a crowd-sourced barometer for public opinion of Elon Musk and Tesla across key milestones: the reveal of the Tesla Cybertruck, the inauguration of President Joe Biden and Musk’s purchase of Twitter, now known as X.
After charting the volume of posts and average sentiment over time, Prem Kumar applied her analysis to Tesla’s stock prices, to understand how reactions to Musk’s public persona affected Tesla’s bottom line.
To her mild surprise, they tracked closely. Though stock prices dipped briefly at times that correlated to large numbers of negative social media posts, they recovered fairly quickly.
“That’s also part of marketing,” Prem Kumar explained. “Either you get people’s emotional reaction or you make something so absurd and people talk about it—that’s word of mouth.”
Akaka says global marketing firms are increasingly pushing their resources into “sentiment analysis,” which crunches publicly available data to understand how consumer behavior is influenced by current events to tease out the nuances and complexity of a brand’s performance.
“When you dig into the data, the relationship between the brand and the founder, and how strongly the sentiment around the brand and the founder correlated, especially with changes in stock prices, was really interesting,” Akaka said.
Now that she’s refined her technique, Prem Kumar says she’d love to expand her analysis with posts from other social media sites, like X, which require more specialized permissions for data access.
Given her findings, what’s Prem Kumar’s best advice for CEOs?
“Be original,” she said. “If you’re authentic, and you’re doing things that make you happy, I think people will automatically assume you know what you’re doing. They will support you regardless.”