Decoding Customer Experience: A Comparative Analysis of Electric and Internal Combustion Vehicles in the US Market Through Structured Topic Modeling

Amid global environmental challenges, the transition from internal combustion vehicles (ICVs) to electric vehicles (EVs) is a priority for governments and automobile manufacturers. This shift requires a deep understanding of consumer preferences and evolving adoption trends. Existing research highlights critical gaps, such as the lack of comparative studies analyzing EVs and ICVs' consumer-perceived value and their evolution over time, and the limitations of static survey methods - currently predominant but constrained in capturing comprehensive consumer insights. To address these gaps, our study utilizes computational text analytics to analyze 13 years of online customer reviews from two major U.S. automotive websites. Using Structured Topic Modeling (STM), we identified 30 factors (in 14 subcategories) influencing EV customer experiences and 40 factors (in 12 subcategories) for ICV customers. By integrating metadata contexts such as satisfaction levels (rating), review timelines, and predicted author gender, we uncovered patterns in functional and non-functional values driving consumer perceptions. This research advances computational text analytics by 1) introducing enhanced methods for STM quality control, 2) developing a comprehensive framework of factors driving EV and ICV consumer perceptions, and 3) presenting longitudinal insights into these evolving preferences. The findings provide actionable insights for policymakers and industry stakeholders. For the EV market, prioritizing affordability, charging infrastructure, and environmental benefits can accelerate adoption. For ICVs, enhancing highway fuel efficiency, reliability, and advanced safety features can enhance customer loyalty. This study lays the groundwork for customer-focused automotive solutions, bridging theoretical understanding with practical application.