Helping human customer service agents handle customer requests at scale

The Challenge

ERGO, a leading European insurance company, wanted to expand their customer service operations to provide 24/7 coverage—all while reducing costs.

Basic chatbots and IVR systems have been rejected by users for being frustrating and unhelpful. Current mass-market solutions are limited to basic, FAQ-based bots that are unable to handle complex requests and have no connection to back-end systems. Given ERGO’s award-winning chat agents and concerns about data security, none of these solutions worked for them.

The Solution

ERGO deployed Rasa Open Source and Rasa Enterprise, laying the framework to automate over 30% of customer requests before continuing to full automation. The ERGO team identified which skills to prioritize through an extensive analysis of thousands of real customer service conversations. A 12-week training phase with agents further improved the accuracy of the system. The AI immediately suggests suitable answers, dramatically reducing response time and boosting productivity and resource availability.

The Results

The deployment convinced top management to invest in making the AI assistant fully autonomous for selected use cases. In addition, the company wants to deeply integrate the assistant into their live chat system.

With Rasa, we at ERGO are completely rethinking how we build customer support experiences. It's about developing faster and cheaper, and at scale. Rasa helps us to realize these goals.

Gregor Wiest, Head of Innovation ERGO Group AG