ThreatMatrix Rolls Out Cookieless Device Identification

A privacy-conscious public has taken to deleting cookies.

* Some popular antivirus software, such as Norton, deletes cookies as part of scheduled maintenance. * Flash cookies (LSOs) can be deleted at the same time as PC cookies are. * Apple’s iPhone, for one, does not support Flash and Motorola’s Droid has strict settings for browser and Flash cookies. * New browser privacy modes make it harder to identify returning visitors.

“The days of tagging customer PCs in order to identify ‘good’ customers logging into user accounts are numbered, as regulatory privacy concerns and privacy settings in Adobe Flash Player 10.1 give end users explicit control over information downloaded to their PCs using Flash Player,” said Avivah Litan, vice president and distinguished analyst, Gartner Research. “Service providers who depend on Flash to identify client devices (such as PCs) in order to prevent fraud should evaluate and implement alternative technologies.” (“Privacy Collides with Fraud Detection and Crumbles Flash Cookies,” Avivah Litan, Gartner Research, February 8, 2010.)

“ThreatMetrix is redefining the device identification landscape with the Fall 2010 release of the ThreatMetrix Fraud Network,” said Reed Taussig, president and CEO, ThreatMetrix. “Relying exclusively on cookies as a means of identifying which devices are involved in an online transaction is no longer adequate to protect customers from recurring fraud or to reliably identify valued returning customers.”

The Fall 2010 release of the ThreatMetrix Fraud Network combines its best of breed ThreatMetrix ExactID™ device identification with ThreatMetrix SmartID technology to detect spoofed devices and IP addresses as well as sniff out botnets. While competing solutions rely primarily on cookies, cookie equivalents that are easily deleted, or a hash of a browser fingerprint that is brittle and continually changes, the Fall 2010 release of the ThreatMetrix Fraud Network provides a more accurate and fraud resistant approach.

“The problem with today’s browser fingerprinting technologies is that they smudge easily,” said Alisdair Faulkner, chief products officer, ThreatMetrix. “Customers change their browser or IP address regularly and fraudsters routinely change their settings. Additionally, botnet scripts randomize device attributes. The result is missed fraud and worse, false positives and declined revenue. ThreatMetrix SmartID overcomes these problems with real-time intelligent matching and confidence scoring.”

ThreatMetrix SmartID complements ThreatMetrix ExactID device identification that positively authenticates customers across PCs and smartphones with the ability to globally track fraudulent devices without false positives. That’s because unlike traditional hashing where every attribute must match exactly to get the same fingerprint, ThreatMetrix SmartID utilizes real-time intelligent matching and confidence scoring to make allowances for changes between visits.

“This enables us to recognize returning devices in much the same way humans recognize faces,” said Taussig. “For example, a witness can still recognize a bank robber based on a distinguishing scar even if they have changed clothes, dyed their hair, or shaved their hair. That’s why ThreatMetrix SmartID is highly resilient whereas traditional device fingerprinting requiring ‘all or nothing matches’ are thrown off when a single feature changes. ThreatMetrix SmartID provides more effective identification of returning visitors that are trying to hide their identity. ThreatMetrix Smart ID can be used by the ThreatMetrix rules engine to identify and correlate suspicious transactions, or it can be fed into third-party identity verification, authorization and authentication solutions.”

The Fall 2010 release of the ThreatMetrix Fraud Network includes several features tailored for financial institutions and high transaction, low latency environments:

Cross Platform Device Identification and Verification