Where technology is strengthening claims
IN Partnership with
Crawford’s Melanie Hughes on building an AI-led model that directs human judgement where it matters
More
Fast track has long been shorthand for speed. Move the file quickly, reduce cycle time, close it, and move on. But that definition is shifting as artificial intelligence moves deeper into claims operations.
“People think about fast track as rushing, as getting it done quickly, and a piece of that is accurate,” says Melanie Hughes, who leads operations for content claims solutions at Crawford & Company following its acquisition of edjuster.
“But when we look at leveraging machine learning and agentic AI, it isn’t so much about speed. Speed is the ultimate outcome, but it’s more about certainty.”
For Hughes, AI’s power lies in giving claims teams a clearer
Crawford & Company is the world’s largest publicly listed independent provider of claims management and outsourcing solutions to carriers, brokers, and corporates, with an extensive global network serving clients in more than 70 countries. The company’s shares are traded on the NYSE under the symbols CRD-A and CRD-B.
“When data is standardized, it becomes pattern-based rather than anecdotal. You’re not relying on someone’s recent experience. You’re looking at patterns across many claims. That reduces leakage, reduces rework, and creates more predictable outcomes”
Melanie Hughes,
Crawford Canada
Share
Share
Copyright © 2026 KM Business Information US, Inc
RSS
Sitemap
Contact us
About us
Conditions of Use
Cookie policy
Privacy policy
Terms & conditions
People
Advertise
Regular Contributors
Authors
E-Newsletter
Contact Us
Contact Us
Data Hub
Benefits
Cannabis
Catastrophe & Flood
Claims
Construction & Engineering
Cyber
Environmental
E&S
Hospitality
Life & Health
Marine
Motor & Fleet
Non-Profits & Charities
Professional Risks
Programs
Property
SME
Transformation
Travel
Workers Compensation
Specialty
Best in Insurance
Business Strategy
E-mag
Events
Guides
IB Talk
Industry Reports
Insurance Companies
Leaders Network
Legal Advisory Board
Legal Insights
Opinion
Premium Content
Regulatory
Webinars
White papers
Resources
Reinsurance
E&S
TV
Insurance News
Industry News
Columns
Diversity & Inclusion
Mergers & Acquisitions
Legal Advisory Board
Risk Management
News
Copyright © 2026 KM Business Information US, Inc
RSS
Sitemap
Contact us
About us
Conditions of Use
Cookie policy
Privacy policy
Terms & conditions
People
Asia
New Zealand
United Kingdom
Canada
United States
Australia
Copyright © 2026 KM Business Information US, Inc
RSS
Sitemap
Contact us
About us
Conditions of Use
Cookie policy
Privacy policy
Terms & conditions
People
Published March 23, 2026
“If you’re going to put more onus on a policyholder to provide information, the tool has to be intuitive. They shouldn’t have to become the expert. The tool is the expert. It guides them toward the right result”
Melanie Hughes,
Crawford Canada
view from the outset. The emphasis is shifting toward straight-through handling, predictive triage, and exception-based decision-making.
“It gives somebody the tools to confidently look at a claim from that 10,000-foot view and know this one really needs more of a human in the loop,” she says. “Versus this one, where we can put the process through and have that human validate. It allows us to leverage human effort in a smarter way.”
Efficiency is no longer defined by speed alone, but by how precisely resources are allocated and how consistently outcomes are delivered.
Smarter triage and straight-through handling AI is changing the starting point of claims handling. Traditional triage prioritized large losses and streamlined smaller claims, but that approach is no longer sufficient.
Severity does not always determine complexity. A modest claim can become time-consuming if information is incomplete or coverage questions arise. A larger claim may proceed smoothly if documentation is clear and expectations are aligned.
AI-driven triage allows claims teams to identify potential friction early. Instead of focusing only on the type or size of a loss, predictive tools can assess whether missing information, inconsistencies, or coverage questions are likely to slow the process. Leveraging that technology surfaces risks and complexities sooner, flagging files for closer review.
The shift toward straight-through handling does not remove adjusters from the process. It changes where their attention is applied. Rather than managing every file end to end, adjusters focus on exceptions, coverage questions, and judgement calls.
Lowering cost while maintaining standards High-volume, low-severity claims can be expensive when each file involves rework caused by incomplete information.
“Where we lose time today is often in the things people don’t
notice,” Hughes says, pointing to “the back and forth” with policyholders who may be first-timers to the claims process.
Structured, AI-enabled intake tools are designed to address that gap. By guiding policyholders through the information required and providing real-time feedback on whether submissions are complete, these tools reduce rework and shorten claim life cycles.
“If you’re going to put more onus on a policyholder to provide information, the tool has to be intuitive,” Hughes says. “They shouldn’t have to become the expert. The tool is the expert. It guides them toward the right result.”
Not every loss requires an in-person inspection. For contained events such as limited water damage, guided capture tools allow policyholders to document losses themselves. AI can then extract relevant data and generate preliminary assessments while flagging anything unusual.
“There will always be claims where you need an expert on site,” Hughes says. “But in others you can put a tool in the policyholder’s hands that guides them through capturing what we need. They can do it on their own time. The technology does most of the heavy lifting.”
Consistency, data, and the human in the loop Perhaps the most durable impact of AI lies in consistency. Variability in how similar claims are documented and assessed has long been a source of dispute. Standardized data capture and valuation tools create a more comparable foundation for decision-making.
“When data is standardized, it becomes pattern-based rather than anecdotal,” Hughes says. “You’re not relying on someone’s recent experience. You’re looking at patterns across many claims. That reduces leakage, reduces rework, and creates more predictable outcomes.”
Maintaining trust in AI-supported processes requires transparency and discipline. Hughes emphasizes the importance of testing tools internally, building feedback loops, and being open about limitations. That collaborative model extends to how teams are trained.
“AI is not perfect,” she says. “You need to acknowledge that. You can’t just accept what it spits out. Build in guardrails so the human using it is comfortable saying, ‘Even though everything looks green, I’m going to take a second look.’”
A different definition of efficient As AI becomes embedded across claims functions, the definition of efficiency is becoming more exact. Speed remains visible, but certainty and consistency are driving
Where AI is delivering the biggest gains in claims
Triage is becoming predictive
Adjuster time is more targeted
Better intake reduces rework
Speed is now an outcome
Consistency improves outcomes
What an AI-led claims model changes for teams
Small claims can require more intervention than large losses
Straight-through handling is expanding; routine files move with minimal manual touchpoints
Coverage, disputes, and exceptions require expertise
Trust requires guardrails
Strong upfront capture shapes the entire life cycle
performance behind the scenes. The question is whether claims teams are spending time where it makes the most difference.
“If you implement AI well, you’ll see a more productive and happier workforce,” Hughes says. “People have more customer interaction and more resolution. They see the value in what they’re doing.”
The redefinition of fast track reflects a broader recalibration in claims handling. Speed still matters. It always will. But certainty, consistency, and targeted intervention are increasingly determining what efficient claims handling looks like at scale.
Contact Us
E-Newsletter
Authors
Regular Contributors
Advertise
IB Markets - Contact Us
Contact Us
Data Hub
Benefits
Cannabis
Catastrophe & Flood
Claims
Construction & Engineering
Cyber
Environmental
E&S
Hospitality
Life & Health
Marine
Motor & Fleet
Non-Profits & Charities
Professional Risks
Programs
Property
SME
Transformation
Travel
Workers Compensation
Specialty
Best in Insurance
Business Strategy
E-mag
Events
Guides
IB Talk
Industry Reports
Insurance Companies
Leaders Network
Legal Advisory Board
Legal Insights
Opinion
Premium Content
Regulatory
Webinars
White papers
Resources
Reinsurance
E&S
TV
Insurance News
Industry News
Columns
Diversity & Inclusion
Mergers & Acquisitions
Legal Advisory Board
Risk Management
News
Asia
New Zealand
United Kingdom
Canada
Australia
United States
Asia
New Zealand
United Kingdom
Canada
Australia
United States