Tracking Consumer Preferences, the Bayesian Way
Greg Allenby

Greg Allenby

The economic downturn hit premium brands hard as consumers opted for lower-price alternatives. This was felt across all industries, from consumer goods manufacturer Procter & Gamble, to full-service insurer Nationwide Insurance, to high-end
retailer Nordstrom.

Greg Allenby, Fisher’s Helen C. Kurtz Chair in Marketing and professor of marketing and statistics, will spend the next year examining whether consumers’ thrifty spending practices will shift back as the economy rebounds.

“The question is whether as we come out of this economic downturn, people’s perceptions of quality are going to return to where they were prior to the downturn,” he says. Allenby is used to tackling tough market questions. He’s a leading practitioner of Bayesian statistics, which lets researchers look in depth at what drives individual purchase decisions. He was recently named co-editor of Quantitative Marketing and Economics, a scientific journal focused on the intersection of marketing, economics and statistics. And he’s the recipient of the 2010 ISMS Long-Term Impact Award for his published research.

“I’m in the business of creating metrics around people’s behavior,” Allenby says, noting that Bayesian statistics excel at providing insights into how and why respondents act in the marketplace. “We can apply this method to all sorts of applied problems. We can deliver a much more precise, accurate analysis for a much tougher class of problems…and move into new arenas of formal analysis not possible before.”

Allenby’s research helps firms understand the critical relationship between needs and wants, benefits that need to be present in brands for them even to be considered for purchase and how advertising, prices and product features work together to produce superior offerings. This ultimately leads to more informed decisions about budget allocation and the best way to increase return on investment.

Bayesians like Allenby recall the days when marketing and other corporate strategy was driven by descriptive analysis based on surveys and case studies. Today, it’s undergoing a transformation in which strategy is increasingly becoming more economically based. Today’s metrics allow firms to merge the richness of survey data responses with observed marketplace behavior.

“That means we can identify the survey variables that have an immediate effect on sales and those with a delayed effect. For the variables with a delayed effect, we can understand how they work to ‘set up’ the sale,” he says, adding that the marketing field, as it’s become more scientifically based, has attracted younger practitioners, better trained in economics and statistical sciences, who are eager to tackle real applied problems.

Allenby consults with a number of firms, including Procter & Gamble, where he’s helping the company better understand consumer concerns around hair care as people age. He also consults with Millward Brown, a leading marketing research agency, and Ohio-based BIGresearch®, which uses Allenby’s forecasting system for its monthly consumer trend study.

Allenby is among five Ohio State faculty researchers working with a new joint venture, Nationwide Center for Advanced Customer Insights.

“NCACI is a great example of industry-academic interaction,” Allenby says. “It’s an example of what we hope to do more of in the future.”

Greg Allenby Interview

Greg Allenby, the Helen C. Kurtz Chair in Marketing, discusses how Bayesian statistics allow marketers to understand consumer behavior.

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

No comments

Show View and Add Comments