Sas wp enterrprise fraud management

WHITE PAPER
Enterprisewide Fraud Management
Analytically powered best practices for detecting, preventing and investigating
fraud in financial institutions

i
ENTERPRISEWIDE FRAUD MANAGEMENT
Table of Contents
Introduction......................................................................................................1
Outwitting the Criminal Mind – The Challenges..............................................3
Bank Fraud Is Increasing in Volume and Sophistication ..............................3
The Fastest-Growing Channels Are Also the Ones Most at Risk...................3
Fraud Is Usually Managed in Business Unit Silos.........................................4
Cross-Channel Fraud Is Widespread.............................................................4
Fraud Detection Has Been Sketchy..............................................................5
Slow Detection Leads to Higher Losses.......................................................5
The Trouble with Traditional Fraud Management...........................................6
Best Practices in Fraud Management – Combating Fraud with the
Technology Available Today.............................................................................7
Step 1: Create an Enterprisewide View of Patterns and Perpetrators..........7
Step 2: Prevent and Detect Fraud in an Enterprisewide Context..................7
Step 3: Investigate and Resolve Fraud in an Integrated Environment........11
Future Trends in Enterprise Fraud Management...........................................11
Closing Thoughts............................................................................................12
About SAS.......................................................................................................14

ii
Ellen Joyner-Roberson, Fraud and Financial Crimes Marketing Manager for
SAS, is responsible for defining industry strategy for the global banking segment.
She has 25 years of experience in the application of information technology to
solve customer needs, including a focus on the financial services industry for
more than 20 years.

Introduction
Sue was just buying pillowcases, paying with the store’s own credit card. But
within days, that card number would reach a fraudster in Florida, who would punch
out new plastic with her number – later, it would be used to buy dozens of store gift
cards at a time, each for an amount just under the review limit.
The debit card came back to the table in no time, and Robert signed the tab for
dinner. He made sure only the last four digits of his card number were displayed on
the merchant’s copy. But Robert didn’t know that in the 30 seconds the waiter had
the card, he had photographed both sides with his cell phone – and would later
use the account to order merchandise and concert tickets by phone.
These events are everyday occurrences – thousands of times a day, actually. In the
first case, the store’s parent company detected the security breach, but only after
three years, 200,000 counterfeit cards and US$1 million in fraudulent purchases.
And if Robert opened his bill to find surprise charges he didn’t make, he was in
very good company; identity fraud is an estimated US$54 billion crime. Some
11.2 million Americans were victimized by card fraud in 2010 – an increase of
12 percent – which was the highest level since the Identity Fraud Survey Report
survey began in 2003, according to Javelin research.1
His bank’s zero liability policy
meant Robert didn’t have to pay the charges, but the bank did.2
1 Kim, Rachel and Monahan, Mary. Javelin Strategy and Research, Identity Fraud Survey Report.
February 2010.
1

Skimming and phishing attacks are becoming more common across the globe.
Recent pay-at-the pump scams have also increased in the US and Europe, and it
requires some help from civilians to fight these types of crimes. No one is safe with the
sophistication of technology and fraudsters today. They can operate globally, within a
matter of seconds, so it’s essential to increase public awareness as financial institutions
enhance their fraud technology and enterprise processes.
1© 2010 The Tower Group, Inc.
352
692
1,189468
580
772
459
556
718
398
575
1,157
0
1,000
2,000
3,000
$4,000
2005 2008 2012P
TRANSACTION MONITORING
CHECK FRAUD
INFORMATION SECURITY
AUTHENTICATION/
ID RESOLUTION
$1,677
$2,403
CAGR = 12.4 %
$3,773
US Banks Will Continue Major IT Investments
in DDA-Related Fraud Management
(USD in Millions)
2008–12
CAGR
19.1%
6.6%
7.4%
14.5%
Source: TowerGroup
Figure 1. According to TowerGroup, investments in DDA-related fraud will continue to
increase, with the largest investment in transactions monitoring capability.
Of course, fraud is not a domestic product. It’s everywhere. Fraud was estimated to
cost the UK economy £38.4 billion (US$62.4 billion) during 2010. This amounts to
£765 for every adult.2
According to the National Fraud Authority’s Annual Fraud Indicator, fraud cost the
UK private sector £12 billion, the public sector £21.2 billion, the charity sector £1.3
billion, and individuals £4 billion in 2010.3
According to Caroline Pearce, the Australian
Payments Clearing Association’s head of fraud, risk and compliance, the industry
needed to “lift our game” to combat a growing criminal threat. The Association found
debit card fraud to have increased 50 percent in fiscal 2010, costing the industry
$0.59 per $1,000 for magnetic strip cards and $0.11 per $1,000 for chip and
PIN technology.4
The good news is that the numbers tend to be slightly down from previous years,
especially in the US. The bad news is that hackers, identity thieves and money
launderers are fighting back by focusing on different channels and spawning new types
of attacks that traditional fraud management strategies were not designed to address.
2 National Fraud Authority. Annual Fraud Indicator. www.attorneygeneral.gov.uk/nfa/WhatAreWeSaying/
NewsRelease/Pages/fraud-costs-the-UK-over-38billion.aspx. Accessed Jan. 27, 2011.
3 National Fraud Authority. Annual Fraud Indicator.
4 Tay, Liz. “Payment Industry Calls for EMV-Compliant ATMs.” IT News for Australian Business.
www.itnews.com.au/News/253292,payment-industry-calls-for-emv-compliant-atms.aspx. April 4, 2011.
2

3
“Interestingly, while FSIs struggle
to capture a full view of each
customer’s relationship with the
institution, enterprising fraudsters
are already achieving this
understanding as they strive to gain
more information about financial
services and products to exploit.”
Rodney Nelsestuen
Senior Analyst, Financial Strategies
and IT Investments, TowerGroup
Outwitting the Criminal Mind – The Challenges
Financial services institutions are well aware of the negative impact of fraud. Even
at industry average levels, fraud hurts an institution’s reputation, customer loyalty,
regulatory compliance and the bottom line. But even the most well-intentioned financial
institutions face some daunting challenges in this area.
Bank Fraud Is Increasing in Volume and Sophistication
“The credit and debit card fraud category of financial services is among the fastest-
growing and best-known means of criminal profit,” says Rodney Nelsestuen of
TowerGroup. “What makes card fraud of great concern is the fact that international
organized crime rings are often involved, turning card fraud from random criminal
activity into industrial-strength enterprises.”5
The sophistication of their tactics makes detecting fraud difficult and preventing it
nearly impossible, especially as the volume of bank transactions grows by about 10
percent a year.
The Fastest-Growing Channels Are Also the Ones Most at Risk
ATM and branch office transaction volumes are flat (as a percent of total transactions),
while call center channels are growing modestly, and electronic access (online and
mobile) is booming. According to research firm IDC, the number of mobile devices –
from smartphones to tablet PCs – accessing the Internet by 2013 will surpass 1 billion,
creating more opportunities for cybercrime.6
Unfortunately, these fastest-growing channels are also the most vulnerable. The
FBI has observed a trend in which cybercriminals — using the compromised online
banking credentials of US businesses — sent unauthorized wire transfers to Chinese
economic and trade companies located near the Russian border. As of April 2011, the
total attempted fraud amounts to approximately $20 million; the actual victim losses
are $11 million.77
5 Nelsestuen, Rodney. TowerGroup. Surrounded by the Enemy: The Case for Enterprise Fraud
Management. March 2008.
6 Shah, Agam. “IDC: 1 Billion Mobile Devices Will Go Online by 2013.” CIO.com.
www.cio.com/article/510440/IDC_1_Billion_Mobile_Devices_Will_Go_Online_By_2013.
Dec. 9, 2009.
7 Fraud Alert Involving Unauthorized Wire Transfers to China. Internet Crime Complaint Center.
April 26, 2011.
“A broad approach will avoid
redundant project work, minimize
system platforms and the number
of vendor partners, increase
staff productivity, and – most
importantly – catch more fraud.”
Gartner, Inc.
Avivah Litan
Gartner Research Enterprise Fraud
and Misuse Management Solutions:
2010 Critical Capabilities
October 2010

Fraud Is Usually Managed in Business Unit Silos
Customers – legitimate and otherwise – tend to see the institution as a single brand
represented across various contact channels: phone, automated contact center, ATM,
branch office and online. But the institution tends to see its customers as diverse
entities based on product: mortgage, credit card, DDA, home equity, consumer
banking, small business, etc. This disparity creates inefficiencies that fraudsters
can exploit.
Cross-Channel Fraud Is Widespread
Criminals know these facts all too well. They know bank fraud systems rarely monitor
customer behavior across multiple accounts, channels and systems. That weakness
opens the door for cross-channel fraud, in which a fraudster gains access to customer
information in one channel and uses that knowledge to commit fraud through another.
“Cross-channel fraud is difficult to discover, track and resolve because the activities
of the fraudster posing as the customer often mirror those of the customer,” says
Nelsestuen. “Thus, apparently normal activity can be fraud conducted over an
extended period.”8
As a result, cross-channel fraud is common – and costly.
Community Mid-Sized Regional
Superregional/
Money Center
All Respondents
Customer victimization scams 3.70 3.82 4.20 3.63 3.83
Debit card (signature based) 3.52 3.60 3.80 3.86 3.63
Check fraud (paper based), deposit 2.91 3.46 3.46 3.19 3.28
Identity theft 3.29 2.96 3.52 3.53 3.19
THREATS AGAINST BANK DEPOSIT ACCOUNTS
(average threat level: 0 = none; 5 = extremely high)
ABA Deposit Account Fraud Survey
Check fraud (paper based), withdrawal 2.73 3.12 3.28 3.25 3.05
Debit card (PIN based) 2.91 2.64 3.20 3.67 2.90
Organized ring attempts 2.21 2.87 3.36 4.07 2.88
New account fraud 2.29 2.65 3.36 3.88 2.79
Check electronification 2.52 2.52 2.92 2.63 2.59
Online banking 2.14 2.33 3.28 3.88 2.58
ACH fraud, originations 2.24 2.28 3.24 3.47 2.53
Wire fraud, originations 2.02 2.45 3.04 3.33 2.51
Remote deposit risk 2.07 2.45 2.71 2.73 2.41
ACH fraud, receiving 2.33 2.21 3.00 2.40 2.38
Wire fraud, receiving 1.61 2.03 2.48 2.20 2.00
© 2009 American Bankers Association. Reprinted with Permission. All Rights Reserved.
Figure 2. All size banks see the highest threats for debit card fraud, while organized
ring attempts and customer victimization scams will require organizations to put fraud
measures in place that monitor fraud attempts across product lines and channels.
8 Nelsestuen, Rodney. TowerGroup. Surrounded by the Enemy: The Case for Enterprise Fraud
Management. March 2008.
4
“EFM enables users to break down
the walls of siloed fraud prevention
operations so that they can look
for fraud and misuse at aggregate
levels, such as at the customer
level, rather than just at an account
or transaction level. This makes
fraud analysts more productive
and versatile in managing fraud
across product lines and channels,
and also uncovers fraud or
misuse that previously would have
been missed.”
Gartner, Inc.
Avivah Litan
MarketScope for Enterprise Fraud
and Misuse Management
January 14, 2011

Fraud Detection Has Been Sketchy
About half the time, fraud is detected by the victims themselves when they review a
monthly statement or are turned down for credit. Imagine the loss of confidence a
consumer would feel to discover fraud before the bank did. Only about 25 percent of
the time does the bank detect the fraud first.
Slow Detection Leads to Higher Losses
Speed is crucial. According to the 2010 Identity Fraud Survey Report, victims who
detected the fraud within 21 hours were defrauded for an average of $373. Victims
who did not discover the fraud up to a month later suffered an average loss of $572.
Those who took up to five months lost nearly three times as much ($1,207) as victims
who detected the fraud within one day.9
This, of course, is no surprise. What can be
a surprise is how much that figure escalates when you add associated costs, such as
lost wages, loss of goodwill and legal fees.
Collectively, these realities create a daunting environment for financial
services institutions:
• They would like to accurately identify the patterns and perpetrators, but they
usually lack the analytical modeling rigor to establish a strong defense.
• They would like to identify cross-channel fraud, but their operational systems
often don’t cooperate well across organizational boundaries.
• They would like to monitor every transaction in real time, but they can’t alienate
customers and merchants with long processing times.
• They would like to implement rigorous rules to detect fraud, but they know they
would turn up a lot of false positives that are costly and fruitless to investigate.
• They would like to unify the fraud management process, but disparate data sources
and cryptic interfaces make the system inaccessible to all but a few people.
9 Kim and Monahan. Javelin. Identity Fraud Survey Report.
5
“EFM also analyzes behavior among
related users, accounts or other
entities, looking for organized
criminal activity, fraud rings,
corruption or misuse.”
Gartner, Inc.
Avivah Litan
Gartner Research Enterprise Fraud
and Misuse Management Solutions:
2010 Critical Capabilities
October 2010

The Trouble with Traditional Fraud Management
To detect fraudulent activity, many banks use transaction monitoring systems – often
homegrown, niche software that requires manual intervention. Still, traditional systems
can work well for detecting individual real-time, point-of-sale fraud. But that’s only one
slice of the fraud pie, and not the biggest slice either.
Few banks have strong, enterprisewide fraud management programs that can correlate
a customer’s behavior across all contact channels and products to identify bust-out
scenarios, social networks and cross-channel fraud. “An institution may have state-
of-the-art security and fraud detection technologies and procedures to protect its
deposit lines of business, but not the same for small business banking or third-party
investments delivered by an alliance partner,” wrote Nelsestuen.10
Furthermore, even the fraud management process itself is fragmented. “Fraud
detection, alert and case management practices are still too often viewed as separate
activities, when in fact they should be managed as a whole,” says Nelsestuen.11
To support real-time monitoring of all types of transactions across all channels – while
adhering to constantly changing regulations such as the authentication guidelines first
established in 2005 – financial institutions must update their fraud prevention practices
along with related fraud technology.
The ideal result would be a fraud management solution that:
• Protects against fraud at the point and time of transaction.
• Accurately detects incidents of fraud in completed transactions.
• Spans all the ways customers interact with the institution.
• Provides structured oversight for the fraud management program.
This is not a halcyon vision. The technology is available today.
Let’s take a look at how a robust, enterprisewide fraud management system can
redefine the economics of fraud. We’ll walk through the process with a hypothetical
bank based on a real one – we’ll call it First Best Practice Bank (FBPB) – a multiservice
institution with more than a million active cardholder accounts.
10 Nelsestuen, Rodney; TowerGroup, Surrounded by the Enemy: The Case for Enterprise Fraud
Management, March 2008.
11 Nelsestuen, TowerGroup, Surrounded by the Enemy.
6
“We estimate that this market
will grow at least 15 percent in
2011, mainly because global FIs
are investing in EFM technology
against a backdrop of increasingly
frequent and sophisticated fraud
attacks, and because they want
to achieve operational efficiencies
by consolidating fraud prevention
and financial crime efforts across
the enterprise.”
Gartner, Inc.
Avivah Litan
January 14, 2011

7
Best Practices in Fraud Management – Combating Fraud
with the Technology Available Today
In 2011, First Best Practice Bank (FBPB) implemented a complete, end-to-end IT
platform for detecting, preventing and investigating both opportunistic and organized
first-party fraud. To begin with, they wanted real-time scoring of all card transactions –
purchase, payment and non-monetary – for faster, more accurate fraud detection on
a global scale. Ultimately, they hoped to prevent fraud before it happened, even as the
fraudsters evolved their methods and hid their deeds in obscure relationships.
Step 1: Create an Enterprisewide View of Patterns and Perpetrators
The new system enabled FBPB to create a true enterprisewide view of fraud. The
knowledge base incorporates data from operational/transactional systems across
separate business units, from Human Resources and audit records, even from external
data sources such as fraud consortium databases. Integrated data quality routines
cleanse and validate the data.
FBPB customized the system with fraud models unique to the institution. Since there
were no preconfigured system limits on rules, they created a deep set of complex rules
for identifying potential fraud.
Step 2: Prevent and Detect Fraud in an Enterprisewide Context
Analytics reveal potentially suspicious patterns and transactions
FBPB’s fraud system goes beyond the typical customer view to provide a holistic view
of fraudulent activity – including related perpetrators and unrelated channels – and
enables a much deeper understanding of customer behavior.
Each transaction – account opening, ATM access, online banking transaction, wire,
ACH, call center encounter, etc. – is passed through a set of rules and predictive
models. In real time, the system checks transaction activity against vast, enterprisewide
intelligence about the customer and potentially suspicious behaviors. Is this an
unusually large deposit for this individual? Is this account linked to another known to be
in a suspected fraud ring? Does this entity hold multiple accounts or similar identities in
unusual ways?
Within milliseconds (for most transactions), the system delivers a score that accurately
predicts fraudulent activity – within or across channels. Even though the system can
operate on billions of records, this transaction monitoring doesn’t bog down real-time
decision making and authorization.
“As financial institutions become more
sophisticated in their fight against
fraud, they are going to demand
software solutions that provide more
flexibility in addressing emerging
fraud issues, analyzing transactions
and activities in real or near-real time
and identifying fraudulent activity
that spans diverse data sources and
payment channels.”
Dan Barta
Director, Enterprise Fraud and
Risk Strategy, SAS

8
Every night, FBPB also runs a batch process of existing customer accounts to
detect and investigate existing fraud as well as prevent new fraud. The system
parses the data and creates a complete update of all account holders and their
key linking attributes. Driven by metadata, all records are exhaustively linked based
on combinations of attributes within the data. Then, using statistical techniques,
common entities are identified and collapsed to produce single views of entities within
networks. Discrete bounded networks within the data are also generated, representing
statistically relevant groups of activities and relationships.
An advanced scoring engine uses independent and combined scores based on three
core processes:
• Application scoring based on scorecard-driven models and text analysis.
• Scoring of individual customers and their full histories.
• Scoring of associated networks, including behavioral data (transaction patterns,
network growth rates, activity levels) and other data provided (current/previous
addresses, contact numbers, employers).
Shorthand customer signatures
Customers are not static. Their personal or professional conditions change. A fraud
management system should recognize and work within this dynamic reality. But how
do you capture it? It would be unrealistic to try to assess every data point around a
customer with every transaction. And if you wait 90 days for a quarterly report, the
opportunity to respond to a change of state may have passed.
The answer can be to use customer signatures – sets of data elements that sufficiently
capture a customer’s state. Examples might be dollars spent on airfare, cash
withdrawals from a credit account or transactions over the median for the cluster.
These data elements are established using regression and correlation analysis. They
then can be used in a shorthand method of identifying the customer’s state – and the
associated marketing opportunities or risk.12
12 Tubin, George. TowerGroup. Consumer Banking Fraud Trends: Welcome to the No-Hype Zone.
May 2008.
“Real-time transaction monitoring
is key to mitigating fraud.”
George Tubin
TowerGroup
Consumer Banking Fraud Trends:
Welcome to the No-Hype Zone
May 2008

9
Is real-time transaction monitoring too cumbersome?
Historically, running thousands of transactions an hour through a host of complex rules
could have been a slow process, noticeably delaying authorization. New processors
and processing techniques have radically changed that, so real-time monitoring is
feasible even for billions of records.
For example, a bank with 30 million active cardholders in the US adopted SAS®
Fraud
Management to check every credit card transaction. In the first three months, this bank
reported the following results:
• An 87 percent increase in the number of card transactions and customer
information processed, while reducing mainframe processing overhead
12 percent – resulting in a 53 percent decrease in mainframe processing cost
per data item.
• A 30 percent decrease in the computing resource cost of processing card
transactions flagged as potentially fraudulent.
• A 10 percent increase in efficiency by agents investigating potentially
fraudulent cases.
Furthermore, the bank reduced IT costs by eliminating three software applications that
were no longer needed with their new, integrated fraud management system.
Alerts from multiple systems are aggregated and
systematically managed
FBPB’s fraud management system aggregates and prioritizes alerts from the bank’s
various fraud detection and money laundering tools. Alerts from all these different
systems are correlated to provide a full picture of the risk associated with an account
or relationship. Analytics can be applied to the alerts to post-score, determine how to
route alerts and support other case management decisions.
An alert database stores the alerts and the results of their disposition. This database
itself becomes a valuable resource for fraud management in several ways:
• Analysts can mine this alert data to determine how well fraud management
tools are generating and feeding the alert management system and to better
understand the nature of false positives.
• The database can automatically communicate with the bank’s host systems
about interventions that have taken place, such as holds placed on funds’
availability, account closure or other actions.
• Performance management systems can use the statistics and metrics associated
with alerts to assess the success of fraud detection and prevention initiatives.
“ In many cases, the lack of
transparency into the risks a
company takes is due to poor
operational control and reporting. …
Technologies are available that can
help overcome these challenges. …
These technologies [include] flexible,
broader approaches such as SAS’
enterprise case management.”
Rodney Nelsestuen
TowerGroup
The Global Imperative for Better
Operational Risk Management: EU
Financial Industry Leads the Way
May 2, 2011

10
Relationships among accounts and entities become clear in
graphic displays
A unique network visualization interface generates a diagram of social network
connections to uncover previously unknown relationships and enable more effective
and efficient investigations. Delving into diagrams of social networks, investigators
can see a complete picture of individual customers, their products, transactions and
connections at the click of a button. They can then drill down into underlying data for
full customer details, including other linked customers and networks. Back up at a
higher level, they can explore entire networks of identities, accounts and applications –
in minutes rather than hours – and take action quickly.
Copyright © 2009, SAS Institute Inc. All rights reserved.
1
Advanced Analytics Lab
Figure 3. SAS®
Fraud Network Analysis allows organizations to use analytics and
network visualization capabilities to help identify key hidden relationships, such as where
organized crime rings are emerging and whether an employee is involved in an internal
fraud scheme.
2
Advanced Analytics Lab
Figure 4. Using dashboards, case managers and executives can check the status of
investigations and prioritize workloads to be more effective and efficient with resources.

11
Step 3: Investigate and Resolve Fraud in an Integrated Environment
Entities and transactions that have been flagged for review are automatically sent to
an integrated case management component. Suspicious cases are automatically
assigned to an investigator. When an investigator logs into the system, he or she is
presented with a list of tasks and a structured environment in which to manage them.
Information is entered in smartforms that are dynamically linked to workflows. The
investigator can:
• See active and pending tasks in context in a visual workflow diagram.
• Display details by case.
• Add freeform notes to the case diary.
• Auto-generate letters from predefined templates and customer
database information.
• Generate summary and detail reports on demand.
The case management component is modular, so it is easy to update as regulations
and circumstances change.
Investigators know their time will not be wasted. Between real-time transaction
monitoring and batch processing, FBPB now sees far fewer false positives. Whereas
traditional approaches yielded one accurate fraud hit for every 30 cases referred for
investigation, the new solution can accurately identify one instance of fraud in every
three – a big improvement.
The case management component also measures productivity and other information
to help direct the fraud management function more effectively. For FBPB, the
automated system with network visualization has reduced the time and effort of
investigating organized fraud by 50 percent to 66 percent. A securities firm reported
that the increase in productivity has enabled them to conduct the same volume of
investigative activity with 26 percent fewer work-hours.
Future Trends in Enterprise Fraud Management
To more effectively prevent future losses, fraud management systems will have
to become self-learning, adaptable to a dynamic environment and evolving
fraud techniques.
Financial institutions already can seamlessly test the effectiveness of fraud-detection
rules and models – and update them when test reports indicate the need. Ideally
though, the system would automatically capture the outcomes of investigations and
reuse those outcomes in future scoring. Models would thereby adapt readily to new
knowledge and continually be refined. Auto-generated network diagrams would
enable strategists to see patterns and symptoms that lead to improved controls and
new monitoring techniques.
“Case management not only provides
a tool for corporate security to
record losses and develop cases
for civil and criminal litigation; it also
provides a repository for detailed
information about fraud exposure
that is essential for maximizing the
effectiveness of fraud detection tools
across the organization.”
Dan Barta
Director, Enterprise Fraud and
Risk Strategy, SAS
“Banks are increasingly looking
to buy EFM and financial crime
software that helps them manage
financial regulatory programs from
the same vendor.”
Gartner, Inc.
Avivah Litan
January 14, 2011

12
This combination of adaptation and visibility would enable financial institutions to better
understand emerging threats, so they can take action to prevent substantial losses
before they happen.
Moving into 2012 and beyond, financial services organizations are looking to use
critical data and information that seem related when working fraud and anti-money
laundering cases. In December 2010, the Association of Certified Anti-Money
Laundering Specialists (ACAMS) and Ernst Young published a joint survey of 250
global respondents from financial services, law enforcement and regulatory agencies.
The survey topic was “Linking anti-fraud and anti-money laundering programs –
unrealized opportunity or unnecessary complexity?” It noted that 52 percent of financial
services firms had integrated some aspects of their anti-money laundering and anti-
fraud functions. The perfect storm of heightened regulatory pressures and increasingly
sophisticated fraud attacks has created an opportunity for fraud and compliance
departments to pool budgets and resources to reduce a shared exposure –
reputation risk.13
1
Trend is moving integration of data and analytics
upstream in the fraud management process
Trend in Fraud Detection - Upstream Integration
Data
Data
Data
Data
Alert
Management
Case
Management
SAS®
Business Analytics Platform
Data Access
Integration
Data
Analysis
Detection
Figure 5. The current trend is to move data and analytics more upstream in the fraud
detection process.
The prospects for the future also extend beyond the scope of any single enterprise. As
more organizations adopt integrated, automated fraud and financial crimes systems,
the potential is there to create a broad consortium of financial institutions that can draw
on their collective experiences to improve fraud detection across the industry.
Closing Thoughts
Organizations that respond to regulatory pressures by simply documenting their
existing fraud management practices are selling themselves short. This is the
opportunity to turn the tide on fraud, fighting back with powerful analytics, holistic
intelligence and integrated case management.
SAS recommends an enterprisewide approach to fraud management that spans all
contact channels and account types. A best-practice fraud management approach is
integrated from end to end, including:
13 Stewart, David. “Financial Crimes Convergence: Tipping Point.” sascom®
voices blog. March 2011.

13
• Data analysis and alert generation: The ability to assimilate data from multiple
sources and apply predictive analytics to accurately assess transactions, activities
and customer state in real time.
• Alert management: The mechanism for accepting, prioritizing and distributing
alerts from the various fraud detection and money laundering tools used across
the enterprise.
• Case management: A structured environment in which to manage
investigation workflows, document loss incidents and report on fraud
management performance.
The technology to implement this approach is available today. The right platform will:
• Integrate with the bank’s existing cardholder and authorization systems.
• Create and manage “signatures” that identify an account holder’s total
behavioral profile.
• Use sophisticated analytic models and business rules to perform on-demand
scoring.
• Make information and alerts immediately available to the people who need
this information.
• Provide a structured environment to manage investigations and
track performance.
The benefits of this approach are substantial. A financial institution could:
• Gain a holistic view of fraudulent activity, including related perpetrators and cross-
channel fraud, and gain a much clearer understanding of customer behavior.
• Improve investigator efficiency with unique network visualization, data drilldown
and other investigation tools.
• Increase ROI per investigator through fewer false positives, prioritization of higher-
value networks and more accurate investigations.
• Prevent future fraud by better understanding emerging threats and taking the right
proactive action.
• Extend the value of the fraud management solution by using it to prioritize alerts
for anti-money laundering, credit risk and marketing applications.
Losses stopped. Fraud avoided. Time saved. The ROI comes from many directions.
TowerGroup estimates that for every dollar spent on fraud management, the enterprise
gains back as much as $8. If enterprisewide fraud management sounds like a good
answer for your financial services institution, flag it for investigation.
“According to The Faces of Fraud
report, 45 percent of businesses
cite loss of productivity as their top
nonfinancial fraud loss.”
Information Security Media Group
2010 Survey Results: The Faces
of Fraud: Fighting Back
December 2010

14
About SAS
SAS provides the capabilities described in this document based on the proven SAS
Enterprise Financial Crimes Framework for Banking. This solution takes a unique
approach that blends multiple alerting techniques in both real time and batch and
provides a systematic framework for investigating and managing fraud cases. A hybrid
solution, it detects and prevents both opportunistic and professional/organized fraud –
including emerging threats such as ACH, wire and mobile. Financial institutions can
significantly reduce losses by detecting more fraud with fewer investigators.
SAS has worked closely with top financial institutions for 35 years to create solutions
to address critical business needs. In the financial services industry alone, SAS data
integration, fraud detection, risk management, regulatory compliance, CRM and
other software is used by more than 3,100 financial institutions worldwide, including
97 percent of banks in the Fortune Global 500®
. Our award-winning solutions handle
the challenges specifically associated with the volatile financial services industry,
and we can help institutions better manage their strategy, risk, customers and
channels to maximize profitability, achieve greater shareholder value and gain a clear
competitive advantage.
SAS is the leader in business analytics software and services, and the largest
independent vendor in the business intelligence market. Through innovative solutions
delivered within an integrated framework, SAS helps customers at more than 50,000
sites improve performance and deliver value by making better decisions faster. Since
1976 SAS has been giving customers around the world THE POWER TO KNOW®
.

SAS Institute Inc. World Headquarters +1 919 677 8000
To contact your local SAS office, please visit: www.sas.com/offices
SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA
and other countries. ® indicates USA registration. Other brand and product names are trademarks of their respective companies.
Copyright © 2011, SAS Institute Inc. All rights reserved. 104003_S70639.0611

Sas wp enterrprise fraud management

More Related Content

What's hot (16)

Viewers also liked (7)

Similar to Sas wp enterrprise fraud management (20)

Recently uploaded (20)

Sas wp enterrprise fraud management