Evolving Consumer Responses to Social Issue Campaigns: A Data-Mining Case of COVID-19 Ads on YouTube
Social issue advertising campaigns are proliferating, particularly on social media and with, in many cases, a polarizing effect.
Yang Feng, University of Florida College of Journalism and Communications Advertising associate professor in artificial intelligence for the Consortium on Trust in Media, and Advertising Associate Professor Huan Chen, wanted to explore ways to monitor online public opinion of the social issue campaigns.
Their new study proposes a new approach to data-mining user feedback considering the dynamic and evolving aspect of comments ranked on social media platforms such as YouTube. The approach identifies top-ranked comments on a social issue campaign and then retrieves popular opinion from the top-ranked comments from a longitudinal perspective.
Five categories of COVID-19 brand videos served as a case study for the authors to test the data-mining strategy and determine which messaging or framing of pandemic commercials the public preferred. The frames included employee appreciation, donation, call to action, frontline worker appreciation, and brand promotion. The researchers concluded that the donation, frontline worker appreciation, and brand promotion frames were favorably received, while the employee appreciation and call-to-action frames were not. The study notes that this could indicate a misperception of the videos’ intentions by the audience, meaning they missed the mark.
While more and more brands are attempting to capitalize on social responsibility and crises issue advertising, they need to be mindful of the potential for backlash. According the researchers, it’s critical to be able to track comments despite the platform’s algorithmic and nuanced ranking system. In addition, the authors contend that future research can apply their newly devised approach to detect the issues consumers are discussing and decide whether the comments echo the advertiser’s desired outcome.
According to the study’s authors, the data-mining approach can also be applied to sentiment analysis of comments and to better understand the characteristics of online commenters, such as through demographic data or tracked online activity. Finally, a study over an extended period could highlight how responses to social issue advertisements evolve as a crisis changes and public opinion shifts from one end of the spectrum to the other.
The original article, “Evolving Consumer Responses to Social Issue Campaigns: A Data-Mining Case of COVID-19 Ads on YouTube,” appeared in the Journal of Interactive Advertising, Volume 22, 2022 – Issue 2.
Authors: Yang Feng, Huan Chen
This summary was written by Dana Hackley, Ph.D.