Evaluating the Effectiveness of AI-Driven Personalization in Enhancing Customer Engagement on E-Commerce Platforms

Abstract

The rapid proliferation of e-commerce has intensified competition, compelling platforms to adopt sophisticated technologies to retain consumer attention. Artificial Intelligence (AI)-driven personalization has emerged as a primary strategy for tailoring user experiences. This study evaluates the effectiveness of such personalization techniques on customer engagement within the specific geographical and cultural context of Pune, Maharashtra, a prominent Indian technology hub. The research aims to assess the direct impact of AI-driven personalization (comprising product recommendations and content customization) on customer engagement (both cognitive and affective) and to investigate the moderating role of perceived privacy risks on this relationship. A quantitative methodology was employed, utilizing a structured questionnaire administered to 363 e-commerce consumers in Pune, selected through convenience sampling. The data were analyzed using descriptive statistics, reliability analysis (Cronbach's Alpha), multiple regression, and moderated regression analysis. The findings reveals that AI-driven personalization has a statistically significant and positive impact on both cognitive and affective dimensions of customer engagement, supporting the first hypothesis. However, the analysis for the second hypothesis confirms that perceived privacy risk significantly and negatively moderates this relationship. The positive effect of personalization on engagement is notably weaker for consumers who exhibit high levels of privacy concerns. This study concludes that while AI personalization is a powerful tool for enhancing customer engagement, its efficacy is contingent upon managing consumer privacy perceptions. E-commerce platforms in Pune must therefore balance personalization with transparent data governance to build trust and maximize engagement. The implications suggest a need for strategies that provide users with greater control over their data, thereby mitigating the dampening effect of privacy concerns.