A contrast-composition-distraction framework to understand product photo background's impact on consumer interest in E-commerce

Mengyue Wang,Xin Li,Yidi Liu,Patrick Chau,Yubo Chen
DOI: https://doi.org/10.1016/j.dss.2023.114124
IF: 6.969
2023-11-16
Decision Support Systems
Abstract:In e-commerce, product photos are a major component of product presentations that aid consumers' understanding of products. In this study, we investigate the impact of the background of product photos on consumers' interest. Drawing upon the attention theories of visual perception, we propose a contrast-composition-distraction framework to understand the product photo background's impact. We conduct an empirical study using a clothing dataset collected from a major fashion product website in China. After differentiating photos' foreground and background and generating features using machine learning, we apply a hierarchical Bayesian model and find that consumers prefer clothing products to be shown on a darker and simpler background. The product should be located in the center of the photo with a slight horizontal offset. It is preferable to use a blurred background and reduce the use of human model faces. These findings are of substantial theoretical and practical value to e-commerce.
computer science, information systems, artificial intelligence,operations research & management science
What problem does this paper attempt to address?