Deciphering the Emotional Code of Employee-Customer Conversations using Voice Analytics
(A2024-119740)

AUTHORS

Saskia Jacob, Karlsruher Intitute of Technology (KIT); Martin Klarmann, Karlsruhe Institute of Technology (KIT); Anne Cordts, Karlsruhe Institute of Technology (KIT)

ABSTRACT

Emotions play an essential role in employee-customer interaction. At the same time, the detection of emotions in conversations can only be done in recent years. Machine learning became a new methodical instrument in this literature field to study input data, that was not useable before. Building on the service and personal selling literature we elaborate the importance of emotions in an employee-customer conversations. With a series of expert interviews (n=10) we find support for their relevance in this context. Building on data from two lab experiments (n=334) we use a self-programmed voice analytics tool to extract dimensional emotion information over the course of the conversation for both conversational partners. Our novel analysis methods provide the possibility to find evidence for emotion contagion on the valence dimension. At the same time, we are able to demonstrate the importance of emotion contagion in a mediated conversation and find significant influence on trust.