Exploring antecedents impacting user satisfaction with voice assistant app: A text mining-based analysis on Alexa services

Abstract

The Amazon Alexa app is one of the most widely used voice assistant apps to manage customer information-seeking behavior and voice shopping. It provides companionship to the visually impaired and senior citizens. There has not been significant empirical evidence focusing on the determinants of user satisfaction with the voice assistant app. Therefore, this study proposed ten user satisfaction antecedents for the Amazon Alexa app based on 13,363 online user reviews. The Latent Dirichlet Allocation (LDA) technique, regression analysis, dominance analysis, and correspondence analysis were used to analyze these reviews. The regression analysis revealed ten user satisfaction predictors of the Amazon Alexa App. The voice shopping experience with the Alexa app aids caregivers in catering to the needs of elderly individuals by providing convenient shopping, voice-activated control, reminders, smart home integration, and access to information. The result of the dominance analysis shows personal assistance and app update are the most important factors for predicting user satisfaction with the VA app. Moreover, personal assistance and voice-controlled automation experience are the most frequent topics for senior citizens and blind people. The findings of this study can provide valuable insights for business managers in determining the prioritization of key determinants of user satisfaction and offering new competitors a competitive edge in the voice assistant market.