The Versicherungskammer Grouphas switched its claims hotline to a speech recognition system provided by Spitch, which understands customer concerns and steers the calls automatically to the right agents. The customers and the insurance company both benefit from this.
The Versicherungskammer Group is part of the Sparkassen-Finanzgruppe – the largest association of financial service providers in Germany. The Versicherungskammer Group is the seventh largest primary insurer in Germany in terms of premium income and employs around 7,000 people. The Group of public insurers ranks second in the German insurance market in terms of premium income.
Until now, the Versicherungskammer Group had used a conventional DTMF IVR with number selection. "The problem with that was that about 40% of all callers either did not make any selection at all or made the wrong one, so they ended up being put through to the wrong agent team”, explains the department manager Miriam Ring. “This caused dissatisfaction among customers as well as unnecessary work interruptions for the agents and additional costs for the insurance company," adds M. Ring.
The solution implemented by Spitch greets every caller with the words "Please tell us about your concerns". The answer, spoken in natural language, is understood by the system, and analysed to determine to which clerk it should be put through. The allocation to the corresponding agent team – liability, motor vehicle, building insurance, etc. – takes place automatically based on the information provided by the caller. The agent sees the transcript on the screen when answering the call, so they know immediately what the call is about. Thanks to the Spitch system, the number of misdirected calls was reduced from 40% to less than 15%. With around 200,000 telephone damage reports per year, 85% of the Versicherungskammer Group customers now reach their destination immediately, save time and are more satisfied.
"This good result was achieved because the Versicherungskammer Group, in coordination with Spitch, repeatedly adapted the language system based on experience from customer behaviour," explains M. Ring: “Natural speech recognition is still new territory for many callers. In a test phase lasting several months, we found out how customers express themselves and what words they use to describe specific concerns, and we adapted the system accordingly. This effort has paid off in terms of the benefits achieved for customers and customer satisfaction as well as the added value for our company.”
The solution is currently being supplemented with new functions and is already performing with a recognition rate of 80%. By the end of the project in the course of 2022, a detection rate of approximately 90% is expected.
Miriam Ring's summary: "We are very satisfied with the solution from Spitch. The project was handled excellently, and the result represents a milestone for us.” The project manager at Spitch, Bernd Martin, explains: “We are very pleased to be able to support one of the most important insurance companies in Germany. Both the course of the project and the result represent an excellent reference for us. We would like to thank the Versicherungskammer Group for that!”
Miriam Ring's video interview and more information about this and many other projects is available via this link.