Persuasion von Konsument:innen durch Chatbots

Abstract

In times of increasing digitalization, the way in which consumers communicate with companies and make purchasing decisions is changing. Digital assistants such as chatbots are increasingly taking on the role of traditional salespeople with the aim of not only providing information but also influencing the behavior of users in a targeted manner. This development raises the question of how chatbots need to be designed in order to be convincing and ultimately contribute to the purchase decision. This master's thesis takes up this question and examines which communicative, technological and contextual factors contribute to the effect of chatbots in the digital purchasing process. Despite the increasing use of chatbots in e-commerce, there has been insufficient research into which features and strategies are actually effective in positively influencing consumers in the decision-making process. This work addresses precisely this research gap and aims to develop a well-founded impact model that explains how chatbots work on a cognitive and emotional level and what conditions are necessary for this effect to lead to a purchase decision. Three established models were used for the theoretical foundation - the Techno-logy Acceptance Model (TAM), the Elaboration Likelihood Model (ELM) and the Theory of Planned Behavior (TPB). Based on these theories, an extended impact model was developed that systematically divides the influencing factors into five categories: Stimuli, Com-municator, Receiver, Contextual Factors and Responses. The methodological basis of the work is a systematic literature review of current studies (from 2020). From this, 19 assumptions were derived that describe the relationship between chatbot characteristics, user perception and consumer behavior. The results show that the quality of communication, the empathic and personalized communication style, the accuracy of information, system availability and the transparent handling of user data in particular are decisive factors influencing trust, credibility and satisfaction. These reactions have a significant influence on the intention to purchase and, if used appropriately, also promote actual purchasing behavior. The developed impact model therefore not only offers a theoretically sound structure for analyzing chatbot communication, but also practical points of reference for the successful use of persuasive systems in digital sales.

Date of Award	2025
Original language	German (Austria)
Awarding Institution	Johannes Kepler University Linz
Supervisor	Wolfgang Jonas Weitzl (Supervisor)