22
1 Copyright © 2010, SAS Institute Inc. All rights reserved. The Insurance Fraud Race Using Information and Analytics to Stay Ahead of Criminals
1
Copyright © 2010, SAS Institute Inc. All rights reserved.
The Insurance Fraud Race Using Information and Analytics to Stay Ahead of Criminals
2
Copyright © 2010, SAS Institute Inc. All rights reserved.
Presenters
Deborah Smallwood Founder, Strategy Meets Action (SMA)
James Ruotolo
Principal for Insurance Fraud Solutions
Global Fraud and Financial Crimes Practice
SAS
© Copyright Strategy Meets Action 2011
- -
Insurance Fraud Race Webinar
An SMA Perspective
April 27, 2011
Founder Strategy Meets Action [email protected] 603.770.9090 Twitter @dmsmallwood
Deb Smallwood
© Copyright Strategy Meets Action 2011
- -
-
An SMA Perspective: The Insurance Fraud Race Today’s Discussion Points
Current State of Fraud
Approaches & Solutions
Business Benefits
SMA Call to Action
4
© Copyright Strategy Meets Action 2011
- -
-
Top 10 Imperatives for Insurers
5
Claims
Analysis
Profitability
Analysis
Risk
Analysis
Unleash
ANALYTICS
On the
DATA
Make GOVERNANCE Work
Make GOVERNANCE Work
Fast Path
NEW PRODUCT
DEVELOPMENT
Drive Dynamic
DISTRIBUTION
Channels
Apply Smarts to
UNDERWRITING
Link CUSTOMER
COMMUNICATION
Holistically
Capitalize on
Intelligence
to manage CLAIMS
Rethink Business
& Technology
LEGACY
Embrace
ENTERPRISE
RISK MANAGEMENT
Chart a Path for
MARKET GROWTH
Source: SMA
© Copyright Strategy Meets Action 2011
- -
-
Current State of Insurance Fraud
An Old Problem in a New Age
Current Efforts to Combat Fraud
Growing Sophistication of Criminals
6
© Copyright Strategy Meets Action 2011
- -
-
Insurer Fraud Fighting Approaches
7
Analytics
and
Advanced
Tools
Information
Sharing • Reporting
• Aggregation
SIUs • Effective referrals
• More skilled resources
Lobbying • Stiffer penalties
• Improved legislation
• More law enforcement
resources
Any Combination…
© Copyright Strategy Meets Action 2011
- -
-
Required Business Capabilities
8
Prevention
Detection
Management
Case Management
© Copyright Strategy Meets Action 2011
- -
-
Current Fraud Management Environment
9
Information Sources
Detection and
Investigation
Tools
Management
Policy Claims Vendors/Other
3rd Parties HR External
Databases
Social
Media Info
Other Unstruc-
tured Data
Manual Case Management Automated Case Management
Spreadsheets
Manual
Analysis
Specialized
Fraud
Systems Custom-Built
Predictive
Models
Physical
Damage
Fraud
Systems Core
Claim
Systems
© Copyright Strategy Meets Action 2011
- -
-
Advanced Fraud Management Environment
10
Policy Claims Vendors/Other
3rd Parties HR External
Databases
Social
Media Info
Other Unstruc-
tured Data
Common Data Repository
Automated Fraud Case Management
Business
Rules
Anomaly
Detection
Predictive
Models
Social
Network
Analysis Integrated Tools
© Copyright Strategy Meets Action 2011
- -
-
Business Benefits from Advanced Fraud Techniques
11
Improve Adjuster/Investigator Efficiency
Accelerated and Enhanced Investigation
Current/Future Saving from Thwarting Organized Rings
Insurer’s consistently report on business benefits beyond loss costs and expense reductions…
© Copyright Strategy Meets Action 2011
- - SMA Call to Action
© Copyright Strategy Meets Action 2011
- -
-
Determine
Your Fraud
Solution,
Create the
Strategy & Plan
Prepare the Data,
Buy Technology,
&
Develop Release
Plan Deliver Quick
Win,
Continue to
Expand & Roll
Out
Strategy Meets Action Call to Action
Find the Champion
Get buy-in
Gain Momentum
13
Source: SMA
© Copyright Strategy Meets Action 2011
- -
-
An SMA Perspective: Featuring as an example: SAS Fraud Framework for Insurance
14
Copyright © 2010 SAS Institute Inc. All rights reserved.
The Insurance Fraud Race Using Information & Analytics to Stay Ahead of Criminals
James D. Ruotolo Principal for Insurance Fraud Solutions
860.633.4338
Twitter @jdruotolo
16
Copyright © 2010, SAS Institute Inc. All rights reserved.
The Shifting Landscape of Insurance Fraud
Insurance fraud is on the rise & today’s schemes are:
Increasingly sophisticated
More agile
Higher velocity
Cross industry
Yesterday’s methods are insufficient
to address today’s fraud risk!
17
Copyright © 2010, SAS Institute Inc. All rights reserved.
Suspicious Claim Identification Methods
Reliance on
rules / red flags
Inconsistent
First-come,
first-served
Advanced
detection methods
Consistent
Optimal
prioritization
vs.
Push Pull
18
Copyright © 2010, SAS Institute Inc. All rights reserved.
…
SAS® Fraud Framework for Insurance
Auto Home Workers’
Comp. General Liability
Life & Health
Detection & Alert
Generation
Real-time
Decisioning
Network Analysis
Alert Management
Case Management
Prevention
Detection
Business Intelligence
Data Quality & Integration
Analytics
SFFI Core Components
Insurance Lines of Business
Business Analytics Framework
19
Copyright © 2010, SAS Institute Inc. All rights reserved.
Proactively applies combination of all 4 approaches at the claim, entity, and network levels
Hybrid Approach
Suitable for known
patterns
Suitable for unknown
patterns
Suitable for complex
patterns
Suitable for associative
link patterns
Policy Claims
Providers Appli-
cations
Payments Referrals
Enterprise Data
NICB
Alerts
Claims History
Rules
Rules to filter
fraudulent claims and
behaviors
Examples:
• Claim within certain
period from policy
inception
• Delay in reporting
claim
• No witness
Anomaly
Detection
Detect individual and
aggregated abnormal
patterns vs. peer groups
Examples:
• Ratio of BI to APD
exceeds norm
• % accidents in off peak
hours exceeds norm
• # claims / year exceeds
norm for policy or
network
Predictive Models
Predictive assessment
against known fraud
cases
Examples:
• Like staged / induced
accident indicators as
known fraud
• Soft tissue injury
patterns across claims
• Like network and claim
growth rate (velocity)
Social Network
Analysis
Knowledge discovery
through associative
link analysis
Examples:
• Claim associated to
known fraud
• Linked policies &
claims with like
suspicious behaviors
• Identity manipulation
SAS Fraud Analytics Using a Hybrid Approach for Fraud Detection
20
Copyright © 2010, SAS Institute Inc. All rights reserved.
SAS® Fraud Framework for Insurance Process Flow
Alert Generation Process
Social
Network
Analysis
Network
Rules
Network
Analytics
Alert
Administration
Business
Rules
Analytics
Anomaly
Detection
Predictive
Modeling
Fraud Data
Staging
Intelligent
Fraud
Repository
Exploratory
Data Analysis &
Transformation
Operational
Data
Sources
Claims
3rd Party
Data
Policy
Enterprise Case
Management
Alert Management &
Reporting
Payments
Learn &
Improve
Cycle
21
Copyright © 2010, SAS Institute Inc. All rights reserved.
SAS® Fraud Framework for Insurance End-to-End Solution
For more information on SAS fraud solutions, please visit www.sas.com/insfraud
Data
• Structured & Unstructured Data
Sources
• Batch or real time
processing
• Data Cleansing
• Data Integration
• Variable Extraction
& Sentiment
Analysis with Text
Mining
Detection
• Business Rules
• Anomaly Detection
• Advanced Predictive
Models
• Watch Lists
• Social Network
Analysis
• Network-level
analytics
• Hybrid Technology
Reporting
• Advanced Ranking
Technology
• Easy to use web
based interface
• Advanced Query of
integrated data
• Full business
intelligence reporting
capability
• Claim system
integration
Administration
• Self administered
• Saas or Installed
• Custom alert queues
• Alert suppression & routing rules
• Workflow analysis
• Direct integration with SAS Enterprise Case Management
Copyright © 2010 SAS Institute Inc. All rights reserved.
