Helping contact center teams improve customer satisfaction
improved answer accuracy with agent assist tool
AI continues to transform the way financial services institutions mitigate risk and create value for their customers. Allianz Direct is at the forefront of this transformation, harnessing cutting-edge technologies to shape the future of online insurance and improve the customer experience. Given their existing relationship with Databricks, Allianz Direct began their generative AI journey on the Databricks Data Intelligence Platform to keep all data and AI workloads in the same environment. The company’s initial use case empowered their contact center agents to spend more time building personal customer relationships that drive lifetime value by handing off mundane agent tasks to AI. As a result of their agents’ trust in the tool, Allianz Direct is now developing new ideas for GenAI solutions powered by Databricks.
Harnessing the power of GenAI to maximize business impact
Allianz Direct is a Munich-based online insurance company that operates as a subsidiary of global insurance leader Allianz Group. On a mission to become “digitally unbeatable,” they are focused on using data to power a superior experience for their customers. The company saw GenAI as a way to enhance customer touchpoints and build personal relationships by helping agents spend less time on back-office tasks. As Des Field Corbett, Chief Technology Officer of Allianz Direct, put it, “We don’t want shorter calls, we’re not trying to replace agents with automation — we’re getting rid of the mundane stuff agents don’t like doing so they can spend more time talking to customers.”
Allianz Direct set up a two-pronged initiative: first, to identify use cases that would deliver real business impact, and second, to evaluate what AI technology would be needed. As an existing Databricks customer, it was an easy transition for Allianz Direct to build their GenAI solutions on the same platform, using Databricks Mosaic AI tools. Corbett explained, “It was pretty much zero effort for us to get started since we had Databricks as our foundation.”
As a technology-driven insurance company, Allianz Direct takes corporate governance and compliance seriously. They recognize the importance of trust in building a relationship with a customer where no physical product is involved. To address questions on governance and ethics, the Allianz Group formed an AI Data Council, and according to Corbett, “While there’s quite a bit of learning still to be done for GenAI governance and compliance, we are making enough progress that it’s not holding us back from seizing opportunities.”
Empowering call center agents with GenAI
For the first Databricks Mosaic AI-powered use case with Allianz Direct, Field Corbett selected an application that could make an impact quickly. The team identified an existing GenAI-powered tool used by their contact center agents when customers had questions about policy terms and conditions. These were common policy questions such as “Can my brother drive my car?” The team built a proof of concept for a RAG-based agent-assist tool. This application incorporated all of Allianz Direct products’ terms and conditions, intelligently delivering the right information to the agent at the right time. The proof of concept was tested with a subset of agents who were eager to apply a solution that would help them better assist customers.
From a compliance perspective, one of the advantages of the agent assist use case was that the company was using publicly available terms and conditions, ensuring the data was shareable. Additionally, the feedback went to a human agent rather than directly to the customer, which provided further confidence. Field Corbett’s team was able to execute the proof of concept much faster with Databricks. He stated, “We used a Databricks Notebook with the steps, and it’s very simple for our developers to make workflow changes.” Compared to the previous GenAI solution, Field Corbett noted that Databricks streamlined both the development and implementation process.
The Databricks team was embedded in Allianz Direct’s Slack channels, engaging with their engineers, helping resolve any issues and further speeding the implementation of the agent assist use case. Field Corbett said, “We made our platform decision based on the working relationship that we had with the Databricks team, who have helped point us in the right direction with GenAI.”
Improving customer satisfaction and unlocking new possibilities with Mosaic AI
The results of the proof of concept have been positive from several perspectives. Compared with the previous application, the Databricks Mosaic AI Agent Framework-powered use case demonstrated a 10 to 15% uplift in accuracy. But that wasn’t the big news. The results led to trust from the agents that this tool was truly helping them. Instead of searching multiple systems and second-guessing answers, the GenAI agent assist tool quickly delivered accurate information to them. This led the agents to use it more often and to suggest additional use cases to make their jobs easier and spend more time with customers.
As Field Corbett’s team rolls the proof of concept out across all Allianz Direct contact centers, he sees a future where GenAI models will give context to an agent before they even talk to the customer. When the customer calls, the agent will already see why they are likely calling, they’ll know the context of the call and will be able to help much more quickly.
Looking ahead, Field Corbett’s team envisions not one or two big use cases, but rather a flood of ways GenAI can impact the business. Databricks has helped Allianz Direct achieve this by supporting multiple GenAI projects with short timelines at a low cost, so Allianz Direct can pick the right models for the right use cases. One of the reasons Allianz Direct originally chose the Databricks Data Intelligence Platform was because of the scalability of its lakehouse architecture and self-service capabilities. Thanks to Databricks, everyone in Allianz Direct has access to the data they need to do their job, securely governed via Unity Catalog. This scalability now applies to GenAI, and Field Corbett’s goal is to empower even more business users to adopt its capabilities across the company.
