How Amwins Transformed Group Benefits Underwriting with AI-Powered Document Extraction

About
AMWINS
Amwins is the largest wholesale distributor of specialty insurance products and services in the U.S.Amwins has expertise across a diversified mix of property, casualty, group benefits and reinsurance products. We also offer value-added services to support these products, including product development, underwriting, premium and claims administration and actuarial services.
Problem
Amwins serves as benefits program manager for one of the largest Professional Employment Organizations (PEOs) in the United States, generating over 8,000 SMB group medical insurance quotes annually. A core feature of our underwriting process is providing detailed, defensible "apples-to-apples" pricing comparisons between a group'scurrent medical coverage and our proposed replacement coverage. For years, this critical capability relied on a manual, time-intensive process: underwriting staff reviewing PDF carrier and TPA invoice documents and manually copying subscriber details into our census tools. This reconciliation step consumed 15-30% of total quote time—anywhere from 20 minutes to 2 hours per quote, depending on group size and complexity. With quote volume increasing 30% year-over-year and no corresponding headcount growth, this bottleneck was becoming unsustainable.
Technical challenges
Solution
Upstage.ai's Information Extraction was the first solution that felt like it had cracked the code on reliable schema extraction from highly variable document sources.The difference? Their document-centric LLM with integrated OCR handled edge cases and novel invoice structures with minimal prompt tuning—better than anything we'd evaluated over several years of searching.Our implementation approach was straightforward:1. Defined our target schema with inline descriptive prompts2. Let the model do the heavy lifting3. Minimal tuning during evaluation yielded 90%+ success rates with 100% accuracy on our test setThe ease of integration into our Azure/M365 platform meant we built and deployed a full MVP in under a month.
Results
Production metrics speak for themselves:
- 1,100+ invoices processed in the first month
- 200+ invoices processed daily by our team of 18 underwriters
- Processing time reduced from 20 minutes-2 hours to under 5 minutes
- 1.5 FTE worth of capacity reclaimed per week across the team
- 3-month payback period on build costs
- 80+ unique carrier invoice formats successfully processed (and counting)
Business impact
The efficiency gains enabled us to:
- Meet SLA commitments without headcount growth
- Absorb 30% YoY volume increases with existing team capacity
- Deliver faster quote turnaround to our PEO client
- Redeploy underwriter time to higher-value activities
What’s next for you?
If your operations are held back by manual processes or unstructured data, Upstage can help. Let’s explore how we can create a solution tailored to your needs. Get in touch today.
Try it yourself
See how Upstage’s Document Parseextracts insights from complex medical records with unmatched accuracy.
Why Upstage
About
AMWINS
Amwins is the largest wholesale distributor of specialty insurance products and services in the U.S.Amwins has expertise across a diversified mix of property, casualty, group benefits and reinsurance products. We also offer value-added services to support these products, including product development, underwriting, premium and claims administration and actuarial services.
Problem
Amwins serves as benefits program manager for one of the largest Professional Employment Organizations (PEOs) in the United States, generating over 8,000 SMB group medical insurance quotes annually. A core feature of our underwriting process is providing detailed, defensible "apples-to-apples" pricing comparisons between a group'scurrent medical coverage and our proposed replacement coverage. For years, this critical capability relied on a manual, time-intensive process: underwriting staff reviewing PDF carrier and TPA invoice documents and manually copying subscriber details into our census tools. This reconciliation step consumed 15-30% of total quote time—anywhere from 20 minutes to 2 hours per quote, depending on group size and complexity. With quote volume increasing 30% year-over-year and no corresponding headcount growth, this bottleneck was becoming unsustainable.
Technical challenges
Solution
Upstage.ai's Information Extraction was the first solution that felt like it had cracked the code on reliable schema extraction from highly variable document sources.The difference? Their document-centric LLM with integrated OCR handled edge cases and novel invoice structures with minimal prompt tuning—better than anything we'd evaluated over several years of searching.Our implementation approach was straightforward:1. Defined our target schema with inline descriptive prompts2. Let the model do the heavy lifting3. Minimal tuning during evaluation yielded 90%+ success rates with 100% accuracy on our test setThe ease of integration into our Azure/M365 platform meant we built and deployed a full MVP in under a month.
Results
Production metrics speak for themselves:
- 1,100+ invoices processed in the first month
- 200+ invoices processed daily by our team of 18 underwriters
- Processing time reduced from 20 minutes-2 hours to under 5 minutes
- 1.5 FTE worth of capacity reclaimed per week across the team
- 3-month payback period on build costs
- 80+ unique carrier invoice formats successfully processed (and counting)
Business impact
The efficiency gains enabled us to:
- Meet SLA commitments without headcount growth
- Absorb 30% YoY volume increases with existing team capacity
- Deliver faster quote turnaround to our PEO client
- Redeploy underwriter time to higher-value activities
What’s next for you?
If your operations are held back by manual processes or unstructured data, Upstage can help. Let’s explore how we can create a solution tailored to your needs. Get in touch today.
Try it yourself
See how Upstage’s Document Parseextracts insights from complex medical records with unmatched accuracy.
Why Upstage
Explore more success stories
Explore how industry leaders have achieved success and transformed their workflows with Upstage.

Automating product price comparisons with a purpose-trained LLM
Upstage collaborated with ConnectWave to develop a domain-specific large language model (LLM) tailored for e-commerce tasks, addressing inefficiencies in manual product data extraction and organization. The customized LLM, built using Upstage’s advanced mixture-of-experts architecture, automated the extraction of attribute-value pairs and streamlined product catalog management. By significantly reducing manual workloads and improving data accuracy, the solution enhanced operational efficiency and user experience. Outperforming general-purpose LLMs like GPT-3.5, Upstage’s tailored model demonstrated the transformative potential of private LLMs in revolutionizing e-commerce operations.

Tax administration made efficient with Upstage Document Parse
Tax work is shifting toward digital transformation and automation. To keep up with these trends, the CEO of a Korean tax firm created a program powered by Upstage Document Parse to boost efficiency. Advanced technologies like Document AI are now essential for handling complex tasks, such as inheritance tax cases, which involve processing large volumes of documents.

Real-time life insurance claims digitization
Building tomorrow’s solutions today
Talk to AI expert to find the best solution for your business.

