The Effects of AI-based Personalization on Consumer Trust, Consumer Satisfaction, and Purchase Intention of E-Commerce Platforms.

Abstract

This study investigates the effects of AI-based personalization on consumer trust, consumer satisfaction, and purchase intention in e-commerce platforms, with a specific focus on the Jaffna District of Sri Lanka. AI-based personalization refers to technologies such as personalized product recommendations, dynamic content delivery, and adaptive user interfaces that tailor shopping experiences based on consumer behavior and preferences. As the adoption of artificial intelligence technologies grows in developing economies, understanding how consumers perceive and respond to these digital interventions becomes crucial for sustainable e-commerce growth. This research is anchored in a combined theoretical framework incorporating the Technology Acceptance Model (TAM) and Expectation-Confirmation Theory (ECT). TAM posits that perceived usefulness and ease of use are key predictors of technology acceptance and trust. Meanwhile, ECT explains how satisfaction is influenced by the confirmation or disconfirmation of prior expectations. Together, these frameworks offer a comprehensive understanding of the consumer trust and consumer satisfaction mechanisms that shape consumer purchase intentions in AI-enhanced environments. A simple random sampling method was used, and the questionnaire was distributed via online platforms such as email and social media among online shoppers in the Jaffna District. Out of 380 responses collected, 347 were validated for analysis. The data were analyzed using SPSS 25 software to examine the effects of the key constructs. The results revealed that AI-based personalization significantly and positively affects both consumer trust (β = 0.42, p < 0.001) and consumer satisfaction (β = 0.38, p < 0.001). Additionally, consumer trust (β = 0.35, p < 0.001) and consumer satisfaction (β = 0.40, p < 0.001) both demonstrated strong, direct effects on purchase intention. These findings support the hypotheses that consumer trust and consumer satisfaction serve as mediators in the relationship between AI-base personalization and consumer purchase intent outcomes. However, the use of simple random sampling and self-reported data may introduce potential biases, limiting the generalizability of the findings beyond the Jaffna District context. This research contributes to existing literature by contextualizing AI-based personalization within a regional Sri Lankan setting. While studies on AI in marketing are growing globally, limited empirical evidence is available for underrepresented areas like Jaffna. Previous Sri Lankan studies have shown that consumer digital adoption is influenced by perceived usefulness and reliability, aligning well with the TAM-based interpretation of this study’s findings. The insights gained from this research emphasize the importance of customer-centric AI design in enhancing user experience, fostering brand trust, and encouraging purchase behavior. The study has practical implications for digital marketers and e-commerce platform developers aiming to increase user engagement in regional markets. Ethical implementation of AI personalization, ensuring transparency, user consent, and data privacy combined with a focus on perceived value and satisfaction can lead to greater consumer loyalty and purchasing outcomes in emerging economies. Future research may consider longitudinal approaches and comparative studies across other districts or cultural contexts within Sri Lanka.