What expectations do human chatbots create in the customer?

What is this blog about?

  • Anthropomorphism is the attribution of human characteristics to non-human objects, such as chatbots.
  • Adding human characteristics to chatbots increases our expectations of what they can do.
  • Several studies show that chatbots with human characteristics are valued more than chatbots without human characteristics.

The research

Crolic and colleagues (2022) conducted several experiments in which they manipulated the human characteristics of the chatbot on the one hand and the moods of the participants on the other. In a first experiment, they created two versions of a chatbot: a version with human characteristics (name, avatar and personal communication style) and a version without these characteristics. Participants first read a scenario that manipulated their mood. In the neutral scenario, participants had to imagine that they wanted to return a defective camera to the company where they purchased the camera. In the angry scenario, the participants read the same scenario, and it was emphasized that they had already tried to contact the company several times in vain. Participants then initiated a conversation with the human or non-human chatbot. They then answered questions about their customer satisfaction.

The results

The results showed that angry participants were less satisfied with the human chatbot than with the non-human chatbot. This difference was not found in participants given the neutral scenario.

A possible explanation for the results is that the angry participants had higher expectations of the human chatbot. To investigate this explanation, Crolic and colleagues conducted a follow-up experiment. In this experiment, the mood of the participants and the chatbot were manipulated in the same way. However, in this experiment, participants answered questions about the chatbot’s service on two occasions: the first after they had just been introduced to the chatbot and the second after they had had the conversation with the chatbot. The results showed that there were no differences between the conditions after the participants had the conversation with the chatbot. However, there were differences in expectations. As expected, participants had higher expectations for the human chatbot. This was especially noticeable in the participants who had been given the evil scenario.

Shortly said:

  • While previous research has shown that human chatbots are more valued, they also raise expectations.
  • Especially angry customers have high expectations of human chatbots.
  • To manage customer expectations, chatbots can indicate in the first messages what they can help customers with. For example, by mentioning specific topics, which the Mediamarkt chatbot does.

Do you know more? This article is based on the publication below:

Crolic, C., Thomaz, F., Hadi, R., & Stephen, AT (2022). Blame the bot: anthropomorphism and anger in customer-chatbot interactions. Journal of Marketing, 86(1), 132-148.

Other references:

Araujo, T. (2018). Living up to the chatbot hype: The impact of anthropomorphic design cues and communicative agency framing on conversational agents and corporate perceptions.Computers in Human Behavior, 85183-189.

Go, E., & Sundar, SS (2019). Humanizing chatbots: The effects of visual, identity, and conversational cues on humanness perceptions. Computers in Human Behavior, 97304-316.

Liebrecht, C., & van der Weegen, E. (2019). Human chatbots: a boon for online customer contact?: The effect of human voice chatbots on social presence and brand attitude. Journal of Communication Science, 47(3), 217-237.

Nass, C., & Moon, Y. (2000). Machines and mindlessness: Social responses to computers. Journal of Social Issues, 56(1), 81-103.

Leave a Comment