FFIEC News

Guardian Analytics Prevents Online Fraud From Login to Logout With FraudMAP Version 2.0

(Aug 19, 2008)-- Guardian Analytics, a provider of online account fraud prevention software for financial services institutions, announced the general availability of a new version of its flagship product, FraudMAP 2.0. This new release extends Guardian Analytics' proprietary Dynamic Account Modeling technology to model all individual online account holder activity session by session, from login to logout, increasing FraudMAP's ability to detect suspicious actions and flag behaviors inconsistent with predicted account holder behavior.

The newly advanced activity modeling capabilities in Guardian Analytics' FraudMAP 2.0 analyzes all online account activity within and across sessions, including seemingly benign actions such as viewing check images or updating contact information. Sophisticated algorithms predict and weigh inherent risk by a number of parameters, such as expected account holder behavior and size of potential loss, to more accurately alert financial institutions to suspicious activity while minimizing false positives. In addition, FraudMAP 2.0 provides banks, credit unions, and brokerages with extensive session and customer-specific context for advanced fraud and identity theft investigation and prevention.

Multi-channel fraud has become a prevalent problem for financial institutions. In these attacks, fraudsters use a compromised online banking account to obtain information such as account numbers and signature blocks to later commit financial fraud via another banking channel, like through the call center or via wire transfer or check writing. These fraud schemes leave no suspicious online financial transactions for account holders or transaction monitoring systems to detect. Moreover, because the actual loss occurs in another channel, financial institutions often mischaracterize the nature of the threat. FraudMAP 2.0 detects these compromises and prevents the fraud in-process before losses can occur.

"Most U.S. banks heeded FFIEC guidance and implemented stronger user authentication for their online banking customers," said Avivah Litan, vice president and distinguished analyst at Gartner. "However, according to our March 2008 survey of 50 U.S. banks, these measures are already being circumvented by thieves using phishing and malware-based intrusion attacks against end users, including cross-channel fraud schemes that typically involve online banking account takeover. Financial institutions should consider using the upcoming 'Red Flag' regulation deadline as an opportunity to improve suboptimal fraud prevention processes cross-account, cross-channel and cross-institution."

The impending Federal Trade Commission's "Red Flag" legislation deadline of November 2008 is placing additional pressure on banks to prevent identity theft with effective prevention programs. FraudMAP 2.0 assists financial institutions in meeting a variety of ongoing compliance, risk and audit requirements. This latest release contains: 1) additional modeling and algorithm enhancements that increase risk scoring accuracy; 2) new user interface modifications that allow fraud analysts to more quickly and easily manage alerts and investigate incidents; and 3) new performance and scalability improvements that meet the needs of larger financial institutions.

"Fraudsters continually demonstrate their ability to compromise traditional approaches to protecting online accounts, with a recent example being the Coreflood Trojan. This scheme steals and validates online banking
credentials, despite the strong authentication and basic fraud detection methods institutions have implemented," said Tom Miltonberger, founder and CEO, Guardian Analytics. "Our new release overcomes the weaknesses of traditional approaches. FraudMAP 2.0's activity modeling, highly accurate risk scoring, and investigative capabilities help financial institutions stay ahead of the fraudsters."

For more information, please visit http://www.guardiananalytics.com.

Share or bookmarklet this web page at: