Intermediate results of the project show (business) value in a number of different ways:

• Providing insights into emotion patterns help in understanding customers and making the interaction with customers services more effective;
• Linking emotion patterns to past service interactions, learning opportunities and successful strategies can be identified to improve service agents’ training;
• The first results indicate that emotion patterns can be used to predict customers satisfaction scores, providing the opportunity to (in the future) eliminate the need for after contact feedback forms to be filled in by customers;
• We show that service managers can benefit from employing an AI model designed to predict stress in order to unobtrusively monitor the stress level of their service agents. Our model has numerous possibilities for practical applications, including real-time early warning systems for service agents, customizing training, and automatically linking stress to customer-related outcomes.

Further research and pilots in commercial call centers are currently taking place to improve and validate the results above.

Published papers

Bromuri, S., Henkel, A. P., Iren, D., & Urovi, V. (2020). Using AI to predict service agent stress from emotion patterns in service interactions. Journal of Service Management.
Henkel, A. P., Bromuri, S., Iren, D., & Urovi, V. (2020). Half human, half machine–augmenting service employees with AI for interpersonal emotion regulation. Journal of Service Management.