AI(A humane AI solution to improve debt collection)

Abakus's AI debt collection platform provides a friendly and humane product solution designed for people who work on the frontline: live agents of the organization. The company's agent training has been enhanced with an AI-friendly culture. Join Ying Liu as she details the results of an experiment showing how the company improved the performance of the collection assistants.

In the online lending business, live agents on the frontline represent the heartbeat of collections organizations. The productivity of the people who connect with consumers on a daily basis has a direct correlation with revenue. Thus, efficient operation management can increase efficiency and performance. This could include anything from improving agent training to creating a strong culture to increase agent retention. Ying Liu focuses specifically on the pain points of Abakus’s AI debt collection platform. A tool engine by a refined model with an area under the curve (AUC) of 0.83 is provided by data scientists, but the assistants in the collections team were always complaining when they found that their experience didn’t match with the AI standards. Ying details the improvements made to the AI tool to truly popularize the product and improve performance, including the combination of human and AI standards, thoughtful design of the autocaller, and content recommendation. More importantly, Ying emphasizes that the performance of the model itself is not everything and that data scientists are supposed to catch the pain points of users in the frontline. We need to make a real difference with AI, and we have the responsibility to create an AI friendly environment for people.