Predicting Consumer Purchase Decisions through Packaging Color Strategies in FMCG Markets

Examining highly competitive Fast-Moving Consumer Goods (FMCG) markets reveals that consumers often make purchase decisions within seconds, rendering packaging design particularly color a critical strategic cue. Although prior research has established the psychological influence of color, few studies have integrated neuromarketing measures with predictive analytics to forecast consumer purchase behavior. This study employs a mixed-methods design combining eye-tracking, physiological emotional measurement (FEMG and GSR), consumer surveys, and machine learning analysis. A total of 1,500 participants evaluated FMCG products across food, beverage, and personal care categories using standardized color treatments. Data were analyzed using ANCOVA and an XGBoost machine learning model to predict purchase decisions. The results show that warm colors significantly reduced time-to-first-fixation (mean = 420 ms) and increased visual engagement, while high-contrast packaging improved fixation duration by up to 32%. Emotional analysis revealed that warm, high-saturation colors generated higher arousal (GSR +18.6%), whereas cooler colors produced stronger positive valence linked to trust. The XGBoost model achieved a prediction accuracy of 89.2%, substantially outperforming traditional regression models. The findings demonstrate that packaging color operates as a neuromarketing stimulus that shapes attention and emotion prior to conscious deliberation. Integrating behavioral science with machine learning advances both theory and practice by enabling accurate prediction of consumer decisions. The study highlights the strategic value of data-driven color design for FMCG marketers seeking competitive advantage in complex retail environments.