The effect of communication styles on customer attitudes: a comparison of human-chatbot versus human-human interactions
(A2019-8393)

AUTHORS

Giulia Pavone, Toulouse School of Management Research-UMR 5303 CNRS University Toulouse 1 Capitole; Lars Meyer-Waarden, Toulouse School of Management Research -CNRS University Toulouse 1 Capitole; Andreas Munzel, Toulouse School of Management Research-CNRS University Toulouse 1 Capitole

KEYWORDS

customer service; chatbot; communication styles

ABSTRACT

Numerous companies are implementing chatbots in customer service, computer programs able to conduct conversations thanks to the natural language understanding and machine learning techniques. Therefore, this study aims to investigate the effect of different communication styles of service agents on costumer’s perceived service quality and attitudes, comparing human-chatbot interactions with human-human interactions. We employ an experimental 2x2 between subject design with two levels of agency (chatbot/human) and two levels of communication styles (social-oriented/task-oriented). If on the one hand results show that a social-oriented style is better perceived than a task-oriented style, on the other hand we haven’t found any difference between attitudes towards the chatbot and the human agent. This result suggests that the theory of computer-are-social-actors may be applied also to virtual agents; therefore, as long as the service is well delivered, customers seem to perceive chatbots and human agents in identical ways.