GermanyFinance and insurance2024 – ongoingAllianz Direct
RAG agent-assist tool for contact center agents
Allianz Direct's contact center agents got answers about policy terms about 15% more accurate with a generative AI assistant
Problem
Contact center agents at the online insurer needed accurate answers about policy terms and conditions while talking to customers. The previous application gave less accurate answers.
Approach
Allianz Direct built a retrieval-augmented generation agent-assist tool on the Databricks platform, developed in a Databricks notebook and governed through Unity Catalog. It started as a proof of concept and was then rolled out across its contact centers.
Results
Takeaway
A tool that answers from the insurer's own policy documents gave agents better answers than the application it replaced.
Categories
- Size
- Large enterprise
- Industry
- Finance and insurance
- Business function
- Customer service
- Innovation type
- Process innovation
- AI technology
- Generative AI
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Databricks describes how Allianz Direct, a Munich-based online insurer, built a RAG agent-assist tool for its contact center agents. The page reports a 15% improvement in answer accuracy and a 10 to 15% uplift over the previous application.
Full details are in the original sources.