Case Study Case Study
Perspectives

Case Studies

Customer Stories

Testimonials

ALM Media, LLC

Josh Gazes, Senior Vice President – Operations

British Gas

Jess Johnson, Head of Operational Excellence

Mosaic Insurance

Mitch Blaser, Co-CEO

Mosaic Insurance

Krishnan Ethirajan, COO

Oxford Nanopore Technologies

Jason Hendrey, Senior Director, Global Customer Services

WS Audiology (WSA)

Christof Steube, Director of Finance Excellence

Kiwi.com

Leonard McCullie, Director, Vendor Management

Kiwi.com

Petra Reiter, Vice President, Customer Services

Flight Centre

Aaron Fadelli, Business Leader

Healius Pathology

Alex Cook, Head of Finance Operations

Varo Bank

Breanna Rivers, Partner Performance Manager

Yorkshire Building Society Group (YBS)

Jessica Lockwood, Process Automation Manager

WS Audiology (WSA)

Sharang Patil, Director of Group Finance Excellence

Priya Madan Mohan, VP for Group Accounting & Controlling

United Airlines

Chris Kenny, VP and Controller

GFG Alliance

Phillip Irish, General Manager, Shared Services Delivery, Quality & Governance

Energy Australia

Steve Corden, Outsource Operations Leader

Delaware North

Christopher Lozipone, Senior Vice President and Global Business Services Head

Moneycorp

Nick Haslehurst, Chief Financial & Operating Officer

Prodigy Finance

Nico Barnard, Head of Operations

M&T Bank

Chris Tolomeo, Senior VP & Head of Banking Services

Minerals Technologies Inc. (MTI)

Khem Balkaran, CIO

Church's Chicken

Louis J. Profumo, CFO & EVP

Smarter Underwriting: Agentic AI-led Research Assistant Accelerates Decision-making

Read | Sep 04, 2025

AUTHOR(s)

A WNS Perspective

Key Points

  • A leading insurer struggled with long research cycles, fragmented data and cognitive overload that slowed underwriting and weakened risk visibility.
  • Leveraging an award-winning multi-agent AI research assistant, WNS deployed a cloud-native solution that blended intelligent orchestration, domain excellence and human oversight to deliver accurate, contextualized insights at scale.
  • The insurer now benefits from accelerated decision-making, enhanced research precision and scalable operational efficiency – creating a foundation for confident action and sustainable growth in a volatile market.
AI Excellence 2025

Business Intelligence Group’s 2025 Artificial Intelligence Excellence Award in the category Augmented Intelligence (Product)

The Industry Landscape
The Complexity of Underwriting Research

Insurance underwriters typically have longer turnaround times for building research reports due to manual workflows and multitude of data sources. Extended validation cycles and outdated records not only delay report delivery but can also skew risk assessments.

Increasingly, advanced analytics and AI-driven interventions are re-shaping this process. They speed up information retrieval, verify data as it arrives and surface insights that were once hidden in disparate sources. By embracing these technologies within a smart framework combining real-time intelligence with human judgment, insurers can move from uncertainty and delay to swift, confident underwriting decisions.

The Client Challenge
Long Hours, Limited Coverage and Bias

The client recognized that its traditional report-building processes inhibited its ability to lead in a fast-moving market.

Operational inefficiency

Operational inefficiency

Hours-long, manual research sessions sapped employee capacity and delayed critical risk insights, leaving the business a step behind emerging trends.

Cumbersome data management

Cumbersome data management

The team struggled to unify data from numerous, often complex sources, creating gaps that undermined proactive risk management. Furthermore, the process relied on ad-hoc approaches that introduced unintended biases.

To sustain competitive advantage and strengthen capital efficiency, the client needed to transform its underwriting research into a streamlined function that could deliver accurate results at speed. This required more than technology; it called for a partner who could thoughtfully combine Al capabilities with domain expertise to craft a smarter, future-ready operating model.

The Solution
The Multi-agent, AI-powered SKENSE Research Assistant

Stepping in as an insurance analytics partner, WNS Analytics deployed a future-ready solution powered by our intelligent solution engineering capabilities and deep domain expertise. Leveraging SKENSE – our AI-driven data extraction and contextualization product – we implemented a sophisticated multi-agent architecture on Azure to transform the underwriting research process.

This award winning, cloud agnostic solution combined a planner agent with specialized execution agents to orchestrate goal-driven research tasks. A human-led oversight layer was integrated to guide, validate and refine AI outputs, ensuring each report met high standards of accuracy, relevance and domain-specific rigor.

SKENSE is an agentic AI-powered product that empowers organizations to tap into the potential of their unstructured data through accurate extraction and contextualization within domain-specific workflows.

Key elements of the solution:

Planner Agent for Structured Research

Planner Agent for Structured Research

  • The agent breaks complex underwriting queries into sequenced sub questions.
  • It ensures end-to-end coverage and citation tracking.
Execution Agents for Data Extraction and Analysis

Execution Agents for Data Extraction and Analysis

  • Web-scraping agent retrieves real-time data from multiple online sources.
  • Document analysis agent processes and validates PDF, HTML and text files.
  • Research query agent creates research queries based on the topic of research for efficient web results.
  • Summarization agent condenses findings into clear, unbiased narratives.
Citation and Validation Module

Citation and Validation Module

  • This module embeds source references in every report.
  • It applies Retrieval Augmented Generation (RAG) architecture with vector database lookup to improve factual accuracy.
User Interface and Feedback Loops

User Interface and Feedback Loops

  • This intuitive portal allows underwriters to submit queries and review draft reports.
  • It allows continuous feedback integration for model refinement.

Key solution features

  • Intelligent data extraction and contextualization

  • AI agents capable of processing multiple document types

  • Gen AI-powered report generation

  • Cloud-agnostic and multi-hyperscaler deployment

  • RAG architecture with vector database

  • Secure, multi-agent orchestration

  • Seamless integration with enterprise applications like Guidewire, Salesforce and Dropbox

  • Adherence to organizational data policies

Tech Stack

  • Azure OpenAI service

  • LanGraph architecture on LangChain

  • Real-time web search and scraping tools

  • 40+ insurance-specific prebuilt AI / ML models with specialized workflows

  • RAG architecture

The Outcome
Accelerated Underwriting and Operational Efficiency

The deployment of this solution reduced cognitive overload for the underwriting teams and significantly accelerated decision making. By combining a step-wise, multi-agent AI architecture with Azure’s scalable infrastructure, our client is now equipped to drive continuous gains in predictive precision, operational agility and sustainable growth.

Key benefits included:

percent reduction in report generation time, with:

No-Loading Icon1
  • ~100,000 research queries processed in a year

  • 20+ diverse data sources integrated per report

percent reduction in research costs, owing to elimination of manual data collection

No-Loading Icon2

percent data relevance and accuracy

No-Loading Icon3