Algorithmic Personalization and Impulse Buying in Social Commerce: The Mediating Role of Hedonic Value under the S-O-R Framework

Abstract

The rapid expansion of social commerce, particularly TikTok Shop, has transformed how consumers encounter and respond to personalized content, raising concerns about the increasing prevalence of impulse buying behavior. Despite this growth, limited empirical research has explicitly examined how algorithmic personalization stimulates affective consumption mechanisms in emerging markets. Drawing on the Stimulus–Organism–Response (S-O-R) framework, this study investigates the effect of algorithmic personalization on impulse buying, with hedonic value as a mediating variable. A quantitative approach was employed using survey data collected from 150 active TikTok Shop users. Data were analyzed using multiple regression and mediation analysis with SPSS. The results show that algorithmic personalization significantly influences hedonic value (? = 0.62, p < 0.001) and impulse buying behavior (? = 0.41, p < 0.001), while hedonic value also has a significant effect on impulse buying (? = 0.47, p < 0.001). The model explains 53% of the variance in impulse buying (R² = 0.53), indicating substantial explanatory power. Mediation analysis confirms that hedonic value partially mediates the relationship between algorithmic personalization and impulse buying (Sobel test, p < 0.001). Classical assumption tests indicate no issues of multicollinearity (VIF < 3), no heteroscedasticity (p > 0.05), and normally distributed residuals. Model feasibility is further supported by a significant F-statistic (p < 0.001), confirming overall model fit. These findings highlight that algorithmically curated content enhances hedonic experiences, which in turn drive spontaneous purchasing behavior. This study extends the S-O-R framework in algorithm-driven commerce and provides actionable insights for optimizing personalization strategies in social commerce platforms.