Frito-Lay study forecasts changing snack preferences

This snack food market is not only growing but also changing as the nation becomes more diverse, educated, overweight and longer-lived – factors highlighted in a new joint study between a UT Dallas School of Management PhD student and Frito-Lay. The study has helped the company adapt its product mix to the changing market for snacks in America.

The research actually started out as the subject of Marina Girju’s summer paper as a first-year PhD student studying management science with a concentration in marketing. At the time, she was interning in Frito-Lay’s Consumer Insights Department, helping analyze data from consumers surveyed about their snacking preferences. Using this data, Girju wrote her paper, which included a forecast model predicting U.S. snacking trends for the next 20 years.

“The paper was about what kinds of snacks consumers eat and how it will change given U.S. Census Bureau predictions. We presented it to Michelle Adams at Frito-Lay, she liked it and gave us some suggestions,” Girju said. “Before, the company knew how dramatic these population changes were, but there was no model that explained whether demographics do indeed influence snack consumption and secondly, how these socio-demographic trends will affect future consumption.”

The summer paper is now a nearly 20-page study that Girju produced in conjunction with co-authors Dr. Brian Ratchford, Davidson Distinguished Professor of Marketing and her adviser; and Michelle Adams, vice president of PepsiCo Customer Strategy and Shopper Insights. The study examines how major demographic shifts might impact snack consumption in the U.S. and how snack-food manufacturers should adjust their business strategies. Using data collected from 30,000 consumers across the country between 2004 and 2010, as well as statistics from the U.S. Census Bureau and other governmental agencies, the co-authors developed a model to predict demand for more than 900 snacks and 80 snack categories in the U.S. market.

The snack consumption forecasting model – known as DemoImpact – captures dozens of demographic variables such as age, gender, marital status, weight and education and uses U.S. Census demographic predictions to help managers understand how consumers are changing and how to prepare for shifts in tastes.

Adams will present key results from the study, “DemoImpact: Modeling, Forecasting and Managing the Impact of Major U.S. Socio-demographic Trends on Multi-Category Snack Consumption,” today at a Marketing Practitioners’ Conference being held at the School of Management. The Frito-Lay executive is one of the marketing experts speaking at the conference and will address how the snack forecasting model has changed the company’s marketing and growth strategies.

The project produced some interesting findings. For instance, regionally:

• Southerners like to snack on cookies, yogurt, chocolate candy, fruits and veggies.
• Midwesterners favor pretzels, popcorn, cookies, chocolates and veggies.
• Northeast residents consume the most pretzels and cookies and the least popcorn.

As for gender, women consume more cookies, yogurt and chocolates, and men eat more potato chips.

Girju, who this fall was named “Outstanding Teaching Assistant” by the School of Management, said the knowledge she gained from conducting this research is priceless.

“The benefits go beyond my first summer paper, beyond helping the company and beyond consumer trends. It started as my first PhD paper, but I never dreamed so much would come out of it,” Girju said. “I learned so much during every step of the process, and everything I’ve learned has helped me grow.”

Provided by University of Texas at Dallas