Faculty

Evaluating TV Ad Effectiveness

Bauer Researcher Develops Framework for Online Behavioral Response

By Julie Bonnin

The Internet has spawned an entirely new platform for advertising, with access to more than half the global population. While it presents nearly endless marketing opportunities, it can also drain advertising dollars as organizations transition through a global digital marketing transformation. Businesses obviously want to utilize the most up-to-date and effective methods amid rapid advances in technology.

C. T. Bauer College of Business Professor Rex Yuxing Du is one of the nation’s foremost experts in using applied analytics to help identify winning marketing strategies. His academic work helps organizations enrich their media planning and evaluate advertising attribution vendors.

The Marketing Science Institute, which shares cutting-edge academic work with the corporate community, frequently promotes Du’s papers and selected him as a member of the inaugural class of MSI Scholars in 2018. In 2017, he was a finalist for the William F. O’Dell Award, which honors a Journal of Marketing Research article that has made the most significant, long-term contributions to marketing theory, methodology or practice. Du has also received Best Paper awards from the American Marketing Association and the Advertising Research Foundation, which have featured his work at main stage presentations at their conferences.

In July, Du and two other researchers published a paper in the Journal of Marketing that presents a framework for evaluating the effectiveness of TV ads based on simultaneous online activity.

The researchers examined a rich dataset consisting of more than 27,000 ad spots on national TV and more than 750,000 spots on local TV for three top pickup truck brands. They then merged that data with minute level search data and found that national advertising spots appear to be more cost effective in generating an immediate brand search response, whereas local spots appear to be more cost effective in generating an immediate price search response. The same framework can be applied for numerous other factors, such as program genres, audience interest levels and more.

“Unlike the proprietary methods used by advertising attribution vendors, our proposed framework for modeling online behavioral response at the minute level is transparent and readily replicable,” Du says. “Advertisers, agencies and networks can use our method as a benchmark in evaluating the proprietary solutions offered by attribution vendors.”

In an earlier study selected as a finalist for the O’Dell Award, one of marketing academia’s most prestigious honors, Du and a co-author presented a statistical model that helps researchers distinguish between longterm trends and passing fads, as they study keyword search data from Google Trends.

“Our research helps marketers see the proverbial big picture in the age of big data,” Du says. “Many firms have adopted our approach. It has helped a lender forecast mortgage default rates, and the demand for different financial products. It’s been used to help automakers identify trends in the importance of different features when people shop for cars. Supermarkets have used it to find out which kind of merchandise is gaining popularity.”

Du is Director of Ph.D. programs at Bauer College and teaches several classes that prepare students to work in Houston’s emerging innovation ecosystem. He received the Bauer Teaching Excellence Award for Outstanding Executive MBA Professor of the Year in 2018.

His work with financial institutions, retailers and some of the largest corporations in the world underscores a motivation to put research to work in the private sector. “For me, the most rewarding moments professionally are when I know my research has made a positive difference in the world of marketing practice,” Du says.