Li Xiao of Fudan University in China and Min Ding, Smeal Professor of Marketing and Innovation at Smeal College of Business at Pennsylvania State University, and advisory professor of marketing at Fudan University said the eigenface method used for face recognition for computer/human security, law enforcement and national security identifies each face by a small set of key dimensions.
Eigenfaces can be considered a set of "standardized face ingredients," derived from statistical analysis of many pictures of faces.
The researchers used the eigenface method to extract and represent facial features in ads with a limited set of eigenface weightings.
In a study involving almost 1,000 participants, the authors used real models’ faces and real ads with minimal modifications to determine the participants' natural reactions to the print ads.
They found different faces change ad effectiveness substantially. In addition, people showed significant differences in their facial preferences across product categories.
The study, published online in Marketing Science ahead of the print edition, found screening faces when designing ads could transform the current subjective process into a scientific one and and increase the number of potential purchasers by as much as 15 percent, and an average of 8 percent.
“This technique will revolutionize the field of ad design,” Ding said in a statement.
This method could substantially increase sales in individual industries. For example, there is a potential for up to $5 billion additional sales for the U.S. automotive industry alone, the researchers wrote in the study.