A faculty member and an alumna from the Naveen Jindal School of Management have won an award from the American Marketing Association Foundation that recognizes researchers whose articles published in the Journal of Marketing Research have made the most significant long-term contribution to marketing theory, methodology, and/or practice.
Dr. Ying Xie, a professor in the Jindal School’s Marketing Area; and Dr. Yiyi Li, PhD’17 an assistant professor of marketing at The University of Texas at Arlington who earned a PhD in Management Science with a Marketing Concentration from the Jindal School; won the 2025 Weitz-Winer-O’Dell Award for their paper, “Is a picture worth a thousand words? An empirical study of image content and social media engagement.”
The team’s research was inspired, Xie, said, by a practical question raised by the founder of a Dallas-based social listening startup company with whom they collaborated in 2015.
“The company aims to help clients such as American Airlines, AT&T and Hilton to improve social media marketing effectiveness,” she said. “Although text-based analytics was starting to gain traction, the increasingly important role of image content due to the proliferation of smartphones with high-quality cameras and the quick rise of Instagram and Pinterest was largely overlooked. The founder also pointed out that firms lacked practical tools to systematically analyze how image characteristics affect user engagement. His observation sparked our interest and motivated us to start this research project.”
The team’s study is one of the earliest attempts to apply machine-learning techniques to systematically analyze image content in marketing.
“Drawing from the extant literature on visual marketing, we proposed three effects of image content on social media engagement: ‘mere presence’ effect, ‘image characteristics’ effect and ‘image-text fit’ effect,” Xie said.
The study provided a practical approach to turn an overwhelming quantity of unstructured image data into structured, quantifiable measures.
“’This approach allowed us to empirically examine how various visual characteristics —such as image quality, colorfulness, human presence, and image-text fit — influence user engagement on social media,” Xie said. “We collected a comprehensive dataset of social media posts related to major airlines and SUV brands from Twitter and Instagram, and our analysis focused on understanding how these visual attributes contribute to user engagement.”
After accounting for selection bias, the team’s findings demonstrated a strong, positive effect of image presence on user engagement on Twitter, consistent across both product categories. They also uncovered several important nuances: high-quality, professionally shot images consistently yield higher engagement across platforms; the influence of colorfulness varies by product category, highlighting the need for context-specific design strategies; and human faces and image-text fit significantly boost engagement on Twitter but not on Instagram, suggesting that platform-specific strategies are essential for maximizing engagement.
“These insights offer practical guidance to marketers, influencers, and content creators seeking to optimize their image content strategies,” Li said. “By understanding which image attributes work best and how these effects vary across platforms, marketers can create more engaging content tailored to their audiences.”
The study’s practical origin and application-oriented focus have made its findings particularly valuable to practitioners, which is why the paper was selected for this year’s Weitz-Winer-O’Dell Award.
“Beyond its practical implications, what we are also proud of is demonstrating that quantifying image content is not only possible but also valuable for deriving insights that were previously inaccessible at scale,” Li said. “While our research was one of the earliest to apply such tools to marketing research, the real contribution lies in highlighting this new possibility and encouraging others to explore and improve upon it. This framework has inspired numerous follow-up studies across multiple disciplines, including Information Systems, Hospitality and Tourism, and Computer Science, which speaks to the interdisciplinary relevance of the paper.”
The long-term impact of the paper is revealed by several key achievements:
- Scholarly Impact: The paper has gathered more than 400 citations since its publication based on the Social Science Citation Index. In addition, the paper’s readership has exceeded 100,000 views and downloads according to Sage journal tracking metrics, demonstrating its broad appeal and influence.
- Pedagogical Influence: The paper has been selected as a required reading in PhD seminars at several top research institutions. Moreover, it has been cited in multiple handbooks on artificial intelligence and image analytics, solidifying its foundational role in this rapidly growing field.
- Practical Significance: The paper has been featured in industry-facing publications. Notably, within a short time since its publication, the documented results have become a part of “best practice” for enhancing engagement on social media.
“What makes this paper especially meaningful is how it bridges theory and practice,” Xie said. “By providing a scalable approach to analyzing image content, our paper offers practical, data-driven guidance for enhancing social media engagement. Its interdisciplinary relevance and adaptability ensure that its contributions will remain valuable for years to come.”
