Document Processing
Legal & General's UK retail retirements division needed to process thousands of pension documents quickly and accurately for more than 5 million customers while meeting regulatory SLA requirements and reducing the manual effort of a legacy scanning process.
Reported impact
Legal & General can process 1,000 documents concurrently in 30 minutes, improved compliance with regulatory SLAs, made the workflow more scalable and cost-effective, and redeployed five employees to higher-value work.
L&G’s UK retail retirements division replaced a legacy document-scanning system with an AWS-based, AI-enabled document ingestion pipeline for pension paperwork. Paper forms arriving from more than 5 million workplace pension members are digitized and processed through a robust, elastic, event-driven architecture that uses Amazon Textract and other AWS AI services to extract and route information quickly and accurately while maintaining service-level compliance. The cloud design scales concurrency independently of staffing, so the operation can process thousands of paper documents each day and handle surges without the fixed limits of the old process. By moving pension-document intake and downstream handling to AWS, the division improved turnaround time, made the workflow more scalable and cost-effective, and freed staff from manual processing so they could be redeployed to higher-value work.
Paper pension documents are digitized and processed through an event-driven AWS workflow that uses Amazon Textract to extract information before routing the data into downstream retirement-operations processes.
L&G’s UK RRD migrated from its legacy scanning solution to AWS, using artificial intelligence (AI)–powered services.
Using AWS, L&G can now process 1,000 documents concurrently in 30 minutes.
Using AWS, we now have a much more robust, elastic, and event-driven architecture.
the division has been able to redeploy five employees, improving business productivity.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
AWS Sample Oct 1, 2024 Intact Financial Corporation It could significantly improve agent efficiency by off-loading repetitive issues to a chatbot and automatically notifying a supervisor when a challenging interaction required assistance. The company sought to build a… 10% calls automatically transcribed within 4–6 hours Amazon Transcribe AWS Sample Jun 19, 2024 Nasdaq Nasdaq needed to help customers reduce the manual effort required for sustainability and environmental, social, and governance disclosures, which meant analyzing large volumes of unstructured documents quickly enough to… AWS Lambda · Amazon Bedrock AWS Sample Jan 1, 2024 TP ICAP When Parameta attended an AWS workshop, it learned how different companies accelerate manual processes and boost productivity by using Amazon Bedrock , which facilitates the way that organizations build and scale… Parameta can then review and confirm the AI-generated assessment. “The compliance team loves the tool’s… Amazon Bedrock 