Paper Type
Short
Paper Number
PACIS2026-1331
Description
Social media marketing increasingly relies on visual content especially consumer-generated images (CGIs) to influence consumer perception and engagement. However, marketers commonly evaluate image performance using engagement metrics such as likes and shares which provide limited diagnostic insight into the intrinsic effectiveness of visual content. This study proposes an AI-driven CGI Evaluation Framework that operationalizes CGI effectiveness as a measurable and multidimensional construct for social media marketing decision support. Using a Design Science Research (DSR) approach, the study develops an artefact that integrates multimodal analytics which include computer vision, optical character recognition (OCR), sentiment-oriented interpretation, and aesthetic analysis to evaluate images across four dimensions: informational, relational, remunerative, and entertainment. The artefact generates normalized scores and radar-based visualizations with explainable recommendations to support managerial interpretation. The artefact design highlights the potential of explainable AI to transform CGI evaluation from reactive engagement tracking into proactive decision support for digital marketing
Recommended Citation
Ibrahim, Siti Zainab, "Designing an AI-Driven Evaluation Matrix for Consumer-Generated Images in Social Media Marketing" (2026). PACIS 2026 Proceedings. 4.
https://aisel.aisnet.org/pacis2026/isdesign_tam/isdesign_tam/4
Designing an AI-Driven Evaluation Matrix for Consumer-Generated Images in Social Media Marketing
Social media marketing increasingly relies on visual content especially consumer-generated images (CGIs) to influence consumer perception and engagement. However, marketers commonly evaluate image performance using engagement metrics such as likes and shares which provide limited diagnostic insight into the intrinsic effectiveness of visual content. This study proposes an AI-driven CGI Evaluation Framework that operationalizes CGI effectiveness as a measurable and multidimensional construct for social media marketing decision support. Using a Design Science Research (DSR) approach, the study develops an artefact that integrates multimodal analytics which include computer vision, optical character recognition (OCR), sentiment-oriented interpretation, and aesthetic analysis to evaluate images across four dimensions: informational, relational, remunerative, and entertainment. The artefact generates normalized scores and radar-based visualizations with explainable recommendations to support managerial interpretation. The artefact design highlights the potential of explainable AI to transform CGI evaluation from reactive engagement tracking into proactive decision support for digital marketing
Comments
13-Design