Smart feedback: Enhancing human feedback and evaluation with AI support in call centers

This project explores the application of artificial intelligence, particularly large language models (LLMs) like ChatGPT, to enhance the feedback process in call centers. Due to the inherent complexity of feedback, we dissect the problem into smaller, manageable components to better understand the effectiveness of these AI techniques. Initially, we assess the performance of LLMs in completing call quality monitoring forms, specifically focusing on dimensions such as warmth and competence of service agents during actual customer interactions. Our goal is to determine how accurately and efficiently these models can evaluate call quality. Next, we examine how human supervisors can leverage these AI tools when filling out call quality monitoring forms themselves. We investigate the potential time savings associated with AI assistance and analyze how this integration affects the evaluations provided. Additionally, we explore the impact of these changes on the feedback process and the quality of the feedback given. Lastly, we assess the attitudes and experiences of both human evaluators and service agents towards the use of these AI techniques. This includes understanding their perceptions of the technology and its impact on their work. By balancing technological advancements with the nuanced needs of human stakeholders, we aim to provide insights into how LLMs can be effectively integrated into the feedback process while maintaining a positive work environment.