- Social media visuals are overtaking text in defining online brand conversations.
- User-generated visuals are reflective of consumer brand perceptions.
- Researchers developed a new model called BrandImageNet to map images to perceptual attributes of a brand.
- Firms can use the BrandImageNet model to automatically monitor consumers' brand perceptions and examine the effectiveness of their positioning strategies.
CATONSVILLE, MD, July 13, 2020 - New research has found that there is a strong link between the visual portrayal of a brand in online imagery created by consumers and the larger brand perceptions.
The research study, to be published in the July/August issue of the INFORMS journal Marketing Science, is titled "Visual Listening In: Extracting Brand Image Portrayed on Social Media" and is authored by Liu Liu of the University of Colorado, Daria Dzyabura of the New Economics School in Russia, and Natalie Mizik of the University of Washington.
According to the authors, their research sought to measure how brands are portrayed on social media and how it relates to brand perceptions. Their goal was to better understand consumer brand perceptions and attitudes toward brands reflected in the imagery consumers post on digital platforms.
"Consumer-created brand images on social media are different from product images on retailer websites," said Liu. "Consumer-created brand imagery posted on social media depicts consumers' interactions with brands and links brands with usage context, feelings, and consumption experiences."
The researchers argued that these consumer-generated images send a powerful message as a form of testimonial for other consumers. They also offer the brand owners the opportunity to understand consumers' brand perceptions.
"In much of the prior research in this area, the focus has been on text content," said Liu. "Given that images are on their way to surpassing text as the medium of choice for online conversations, monitoring visual content is important to get a more complete understanding of online conversations involving brands."
In the process, the researchers introduced a "visual listening in" approach to monitor visual brand content that was created and shared by the actual consumers on social media. They developed and validated a model, BrandImageNet, to allow firms to monitor their brand portrayal on social media and evaluate it relative to competitors' and their own firm's desired brand positioning.
"Our BrandImageNet model maps images to specific perceptual attributes," said Liu. "We focused on identifying perceptual brand attributes rooted in brand images. This is different from identifying functional attributes of the product itself. One example we use in our research is a comparison between the Prada and Eddie Bauer brands. Which one is portrayed by social media users, visually, as the more glamorous and which one is the more rugged one? Our model unequivocally points to Prada as glamorous and Eddie Bauer as rugged. Across all brands in our study, we find a strong link between model predictions and consumer brand perceptions collected with traditional survey-based methods."
About INFORMS and Marketing Science
Marketing Science is a premier peer-reviewed scholarly marketing journal focused on research using quantitative approaches to study all aspects of the interface between consumers and firms. It is published by INFORMS, the leading international association for operations research and analytics professionals. More information is available at http://www.