Examining the Impact of Artificial Intelligence on Optimizing Customer Experience in Online Retail Stores

Abstract

The primary objective of this study was to analyze various dimensions of employing artificial intelligence (AI) to enhance customer experience in online retail stores and to identify the factors influencing its success. This research was conducted using a quantitative, descriptive-survey method, and the relationships among AI, service personalization, and customer experience in online retail were examined through structural equation modeling (PLS-SEM). A purposive sampling method was applied to a group of 400 individuals, and data were collected using a researcher-made questionnaire. The validity and reliability of the instrument were confirmed through statistical analyses, and data analysis was performed using SmartPLS 4 software. The mediating role of personalization was assessed using the bootstrap method, and model fit was evaluated using standard indices The key findings of this study demonstrated that the use of AI has a positive impact on both customer experience and service personalization. The effect of AI usage on customer experience was confirmed with a path coefficient of 0.31 and a significance level of 0.001, while its effect on service personalization was confirmed with a path coefficient of 0.54 and a significance level of 0.002. Furthermore, service personalization showed a positive and significant effect on customer experience. In examining mediation, it was found that service personalization plays a partial mediating role in the relationship between AI usage and customer experience. Model fit indices including SRMR, NFI, and GOF indicated a good fit and the model’s capacity to explain the relationships among the variables. The findings of this research revealed that the integration of AI in online retail stores significantly enhances customer experience. This technology, by strengthening service personalization, directly and indirectly influences customer satisfaction. Additionally, the structural equation modeling analysis confirmed that the proposed conceptual model fits well and effectively identifies the complex relationships among the variables.