AI and machine learning–based systems are increasingly used in retail to forecast promotional demand and support complex promotion planning decisions. Yet many retailers continue to rely heavily on managerial intuition and simple performance indicators, indicating a gap between advanced analytical capabilities and their effective use in decision-making. This conceptual paper develops a socio-technical framework for AI-based promotion forecasting in organized retail that integrates technical and organizational perspectives. Drawing on literature on AI-driven promotion analytics, customer segmentation, decision support systems, human–AI collaboration, and AI governance, the framework theorizes how customer segmentation sophistication and machine learning–based forecasting inputs influence managers’ perceptions of forecast accuracy and decision support quality. It further proposes that interpretability and managerial trust mediate these relationships, while organizational readiness and governance structures moderate the extent to which AI-generated forecasts shape promotion decisions. Based on this integrated perspective, the paper formulates propositions to guide future empirical research and outlines implications for designing AI-enabled promotion forecasting systems that are technically robust, interpretable, and aligned with managerial and organizational realities in retail.