Rethinking Marketing Data Through Visual Analytics
The Problem With Pictures
Why might one marketing image outperform another? Is it the lighting, the presence of a person, or the accompanying caption? I would want to understand those differences before treating a successful image as a formula.
According to Adobe Express (2026), effective marketing graphics depend on several design decisions, including visual hierarchy, color, layout, and readability. However, knowing what makes a graphic visually appealing is different from understanding which elements actually contribute to its performance. This is where image becomes valuable.
These questions connect directly to the challenge that Yiyi Li, Assistant Professor of Marketing at the University of Texas Arlington, and Ying Xie, Professor of Marketing at the University of Texas at Dallas, explore in From Pixels to Market Outcomes: A Framework for Image Analytics in Marketing, published by the American Marketing Association on March 19, 2026.
What the Article Covers
As Li and Xie (2026) explain, analysts must decide which visual characteristics are relevant to their goals and how to measure them consistently (Li & Xie, 2026). Their framework organizes image analytics into three levels; low-level features, such as color and brightness; mid-level features, such as people, objects, and logos; and high-level features, such as emotions and perceived quality.
The Article also discusses how images and accompanying text work together to influence consumer responses. Rather then assuming that certain visuals perform better, marketers should consider the audience, message, and desired outcome when analyzing images.
What This Means for Digital Analytics
Traditionally, marketers might compare images using clicks, engagement rates, or conversions. Image analytics adds another layer by helping explain which visual characteristics may be associated with those results.
For example, if two advertisements receive different engagement rates, marketers could examine their visual features alongside their performance data. However, identifying a relationship does not necessarily prove a particular color, person, or design element caused the difference.
This development connects to what we have learned about using analytics to support marketing decisions. Jackson (2016) emphasizes the importance of connecting measurements to organizational goals. Image analytics could help marketers make more informed creative decisions, provided they evaluate the findings with the context of the campaign.
My Perspective: Data Should Inform Creativity
I consider image analytics a positive development because it creates opportunities to understand how visual communication contributes to marketing performance.
With my interest in both marketing and graphic design, I find this especially relevant. Design involves making creative choices, but those choices should also support the intended message and audience
My concern is that marketers could become too dependent on identifying patterns and begin treating successful design as formulas. Consumer preferences are not always predictable, and the same visual elements may produce different responses depending on the audience or situation.
I believe image analytics works best when it supports creativity rather than replaces it. The goal should not be to make every image look like the highest-performing advertisement, but to understand what makes visual content meaningful and effective.
References
Adobe Express. (2024). Social media graphics: Beginner’s design guide | Adobe. Adobe.Com. https://www.adobe.com/express/learn/blog/design-tips-for-social-media-graphics
Jackson, S. (2016). Cult of Analytics Data Analytics for Marketing (Second Edition). Taylor & Francis.
Li, Y., & Xie, Y. (2026, March 19). From Pixels to Market Outcomes: A Framework for Image Analytics in Marketing. American Marketing Association. https://www.ama.org/marketing-news/from-pixels-to-market-outcomes-a-framework-for-image-analytics-in-marketing/
Leave a Reply