Leveraging AI And A/B Testing to Optimize User Experience and Conversion in E-Commerce

Abstract

With the use of artificial intelligence (AI) and controlled experimentation, optimization of digital user experiences and conversion strategy in e-commerce settings has become a key poster child. This survey describes recent developments and models of A/B testing, multi-armed and contextual bandits and hybrid experiments powered by AIs. Although some of the current methods have shown significant improvements in the performance indicators like click-through-rate and conversion rate, there are still a number of practice and methodological questions to ask. This discussion also provides a critical review of the existing research gaps, such as the scalability problem, the understandability of the models and the adaptive design of the experiments. Models proposed and experimental data offer empirical real-life understanding of the relationships between AI algorithms and empirical validation methods. The review also ends with pointing at the potential future directions of better personalization and adaptive testing limited by the real world.