How a Nordic insurance company automated claims processing

Navigating large volumes of unstructured insurance claims data is a substantial challenge for any insurance company looking to automate repetitive yet complex tasks across its operations. The integration of advanced technologies such as automation and artificial intelligence (AI) has become a pivotal driver for the exponential transformation of insurance organizations.

A leading insurance firm in the Nordics wanted to modernize its claims management process. The current workflow required manual processing of each claim request, a time-consuming series of repetitive tasks which had to be performed by its agents. Amplifying the complexity was unstructured data from various sources, requiring detailed examination and categorization. The spectrum of claims data was also broad and varied, originating from multiple sources such as bills, invoices, cash receipts from pharmacies and local clinics, and medical treatment diagnoses with supporting documents.

Each document required thorough analysis by the agents to extract necessary data that could then be categorized based on the relevance for claim processing. As a result, customer service efficiency suffered; the teams struggled to process claims swiftly due to the limited capabilities of existing technology infrastructure that was ill-equipped to facilitate end-to-end automation.

A transformative approach was necessary. The insurance company engaged with EY teams to analyze its existing tech infrastructure and business needs. The goal was to develop an AI-based solution that could streamline their claims management process, automate routine tasks, and free up their agents to concentrate on building stronger customer relationships. By placing humans at the center of this AI-powered transformation, the company sought to augment the agents' capabilities, enhancing their performance and job satisfaction.

This strategy extended beyond a technological upgrade. It was about empowering their people, preparing them for a future where they can collaborate effectively with AI to drive impactful results. This initiative aimed to set the stage for the insurer's expansion into new markets, with their AI-powered claims management solution acting as a catalyst.

Transforming the insurer's traditional work environment required a robust, AI-powered solution. Manual handling of claim reimbursements involved laborious tasks like opening individual images and PDF scans, analyzing files, and uploading relevant documents to their systems. EY teams stepped in with EY Fabric Document Intelligence, an AI-powered product, to streamline this process.

Developed by EY wavespace™ Madrid and now integrated into EY Fabric, EY Fabric Document Intelligence is a technology that converts semi-structured and unstructured documents into actionable structured data. Built on machine learning and the Python programming language, the tool is hosted on an EY-secured cloud environment. It was scaled and customized to meet the insurer's specific requirements during the project.

The process was simple yet transformative. Agents uploaded the scanned copies to EY Fabric Document Intelligence, which began image cleansing by detecting relevant data. This involved removing image backgrounds, correcting document rotation and reducing noise to enhance the quality of scanned files. The product then performed preprocessing, document analysis and layout analysis. Using optical character recognition (OCR) and natural language processing (NLP), unstructured data was converted and classified. Finally, the structured data was transferred to the insurer's core claims system.

The product processes more documents over time, the system improves and has the potential to provide continuing value and insights for the insurer. Our team provided a solution that did more than just digitize and organize unstructured data. It combined EY business insights and industry knowledge within a strategic technology ecosystem to help modernize the insurer's operations. EY professionals offered comprehensive services, including solution design, system integration, data science, project management and cloud computing knowledge, with a human-centric approach.

Teams from Sweden, Denmark, Spain, the US and the UK collaborated to integrate EY Fabric Document Intelligence with the insurer's legacy system. They helped the insurer understand that AI, when guided by human insight, serves as a powerful enabler to expedite work.

The insurance firm now benefits from near real-time processing of claim documents. Since the solution's implementation, a remarkable 70% of the documents fed into the system are correctly extracted and interpreted. This crucial upgrade not only expedites decision-making but also allows agents to concentrate their efforts on enhancing customer interactions. Agents have more time to spend with their customers in providing personalized advice. This shift toward more meaningful engagements cultivates stronger relationships, maintaining trust and driving additional business value. The solution is designed in a way that it gives the required control of the automation and AI technologies to the insurance firm. The data is entered into the system and the output is driven through a controlled set of confidence levels avoiding blackbox implementation of AI. As the client was closely involved in all decision-making, they have sufficient transparency to the solution.