Are personalized smart ad video display systems the next big advantage for advertisers—and consumers?
In a recent article published in Information Systems Research, Scheller College of Business Information Technology Professor D. J. Wu , the Ernest Scheller Jr. Chair in Innovation, Entrepreneurship, and Commercialization, studied both the feasibility and effectiveness of developing personalized smart ad video display systems.
Wu co-authored the study with Associate Professor of Marketing Li Xiao, Fudan University, and Bard Professor of Marketing Min Ding, Pennsylvania State University, University Park.
“Video ads are increasingly ever-present, showing up in banks, hotels, office buildings, elevators, and even on city streets,” explained Wu. “Our research delves into the next steps necessary to transform that digital signage into smart signage. While current digital signage is definitely an improvement over traditional print signs, most lack a smart component and are not designed to capture consumer ad preferences and offer personalized ads that might elicit positive responses.”
Current Digital Signage
The market for digital signage is big—and growing. Grand View Research (2023) reported that the size of the global digital signage market was estimated to be $24.86 billion in 2022 and is expected to expand at a compound annual growth rate of 8.0% from 2023 to 2030.
In recent years, there has been increasing movement toward smart ad display systems. Some retailers in the U.S. are using in-store video systems with screens and software that display ads tied to the viewer’s gender and age. However, those ads often miss the mark; viewers of the same gender in the same age group will still have distinct differences in their preferences.
In the study, Wu and his colleagues attempt to address this issue and find a way to offer personalization based on individual preferences. (The systems will include an opt-out system to address privacy issues).
“The goal is to select a set of video ads that appeal to individual consumers,” said Wu, “then display them in a way that enhances the consumer’s ad viewing experience and communicates the ad message clearly and effectively.”
How It Will Work
Wu and his colleagues’ proposed system would capture consumers’ facial expressions and eye gaze stream data as they watch an ad and then analyze data at the frame level. The recognized facial expression and detected eye gaze would then be matched to the corresponding frame of the video ad, linking facial expressions to specific visual objects appearing in the ad. By tracking a consumer’s facial expressions in response to various visual objects in real-time, the system will learn the consumer’s individual preferences toward different ads, search the ad pool, and select and subsequently display a new ad that is most likely to elicit positive attitudinal and behavioral responses.
“To put it simply, the proposed system will measure what a person is most drawn to and then use that information to select ads that match the viewer’s preferences,” said Wu. “For example, if a person’s gaze lingers on an ad for food, more food ads will be shown.”
Results
The results of the study are encouraging. By tracking a consumer’s facial responses to only one ad or even part of an ad, the proposed system can make reasonably accurate inferences about a consumer’s ad preferences. Those inferences can then be used to make personalized recommendations that help enhance consumers’ ad-viewing experiences and elicit favorable responses. This represents a win-win for both the advertiser and the consumer/viewer: the advertiser enhances the effectiveness of their advertisement (by targeting potentially interested individuals) and avoids wasting ad resources, while consumers enjoy a more relevant and pleasant experience at no additional cost.
"The study marks a first step towards enabling machines to read people's minds and better serve consumers in the era of generative artificial intelligence," summed up Wu.
To download the entire study, visit Information Systems Research.