An Ounce of Fraud Prevention Goes a Long Way

An Ounce of Fraud Prevention Goes a Long Way

An Ounce of Fraud Prevention Goes a Long Way
Cooper Fraud Score

Emboldened and sophisticated fraudsters are driving an increase in card fraud losses. Financial institutions are taking more aggressive, costly measures to combat it, but not all strategies are equally effective. However, new approaches that blend cutting-edge technology with the use of human consultants are proving to be more effective at preventing fraud while preserving the member experience for credit unions.

Global card fraud losses have tripled in the past decade and are expected to total $408 billion over the next 10 years. The United States is at the center of the crisis; despite accounting for just 22% of worldwide card transaction volume in 2020, U.S. issuers were hit with nearly 36% of global card fraud losses in 2020.

Making matters worse, over a third of surveyed industry fraud executives say that well-organized criminal rings are committing up to three-quarters of such attacks instead of “lone wolf” fraudsters.

A Multi-Layered Approach to Reducing Card Fraud

Credit unions and other issuers face several challenges as they attempt to address the rising tide of card fraud. As fraud losses increase, they lead to rising credit union operational costs—cutting into payment revenue margins and putting pressure on the dividends credit unions return to their members.

Yet, the existing industry-standard fraud prevention methods aren’t keeping up with the times—or with an increasingly sophisticated and organized breed of criminals – and are either too severe or not doing an effective job of catching individual instances of fraud, or worse, both.

With threats coming from all corners of the evolving payments landscape, it’s easy to see why no single solution can safeguard credit unions and their members while protecting member relationships. Yet integrating multiple solutions from multiple providers presents its own set of security and member-experience risks. Co-op has invested in developing a single hub of fraud detection and prevention solutions bolstered by decades of human expertise.

According to Nicole Reyes, Director, Co-op Solutions Fraud Prevention Services, this multi-layered approach is the key to effective fraud prevention. “Fraud is ever-evolving, so we must respond with agility,” Reyes says. “Co-op’s Fraud Prevention & Solutions Consultants are an extension of a credit union’s fraud team, which maintains its subject matter expertise in the distinction between valid and fraudulent spend patterns. Our consultants then collaborate with the credit union fraud team to tailor custom fraud rules. This ensures we are meeting the distinctive needs of each credit union and its membership.”

One important tool of this approach is Co-op Cooper Fraud Score—a machine learning-enabled solution for recognizing and stopping fraud in real time. Cooper Fraud Score also includes reason codes alongside the score to provide context on why a transaction scored a certain way which can provide insights on fraud trends and help when servicing member inquiries on declined transactions.

Cooper Fraud Score is a dynamic, integrated, real-time machine learning score that empowers credit unions to react more quickly to fraud trends while maintaining the highest levels of service that members expect. Importantly, the solution reduces false-positive ratios, which leads to fewer in-person and online declines for legitimate member purchases. Co-op’s fraud team monitors performance and data in real-time on behalf of credit union clients to put new rules in place more quickly. Continually leveraging insights over time enables bi-annual model updates, something that is highly unique within today’s scoring industry.

Beating Fraudsters at their Own Game: Early Success

Co-op recently ran a trial of Cooper Fraud Score, and the results were very positive. The goal was to test its effectiveness as a new real-time decisioning element in production. Cooper Fraud Score was particularly successful in detecting low-dollar amount fraud, a common tactic of fraud rings trying to capture quick wins just below the thresholds of traditional fraud prevention methods. The trial also monitored how Cooper Fraud Score performed in detecting increased transaction velocity and massive BIN attacks, which are other commonly used fraud schemes.

Las Vegas-based WestStar Credit Union participated as a beta for Cooper Fraud Score. “Solving the problem of fraud is as much about meeting member expectations as it is about maintaining a healthy income statement,” said WestStar CEO Rick Schmidt. “It is a massive frustration when a member calls up and asks how we missed an obvious fraud, and we have no good answer for them. Bad guys are really creative. You’re never going to catch it all, and that’s not the expectation, but we should be able to catch the low-hanging fruit. Throughout the beta experience, we saw that Cooper Fraud Score is smarter about preventing that kind of fraud.”

In the trial, Cooper Fraud Score demonstrated an 11% lift in the credit union’s ability to detect instances of fraud, resulting in a reduction of $1.3 million in potential fraud losses.

How Cooper Fraud Score Works

Cooper Fraud Score is designed to work in concert with existing scoring and fraud strategies and in a series of steps:

  1. Establish fraud rules: Co-op’s Fraud Prevention & Solutions Consultants work with the individual credit union to establish initial transaction approval rules based on the credit union’s risk tolerance, cardholder base characteristics, existing fraud patterns, and other factors.
  2. Real-time scoring: Credit and debit transactions are scored in real time, right in the data stream, which allows credit unions to take quick, decisive action on suspicious transaction activity and eliminate fraud losses before they occur.
  3. Fraud detection: Leveraging existing fraud alert and case management tools, suspicious transactions that meet established fraud rules are declined immediately, and a reason code is provided.
  4. Continuous learning: Because Cooper Fraud Score is built on a machine learning-enabled continuous feedback loop, the model can identify anomalies in behavior and react to evolving fraud trends. As transaction decisions and case outcomes are fed back into the model, performance improves over time.
  5. Reporting: Coming soon, new dashboards and current fraud reports will provide high-level and detailed views of transactions scored by Cooper Fraud Score, helping management to identify recurring trends and take further preventative action steps.

“We are excited about Cooper Fraud Score performance in its initial release as a new tool to help prevent fraud in real-time as well as positively impact the member experience,” said Patrice Alexander-Lee, Product Management Director for the Protect line, Co-op Solutions.

As credit unions seek new and better ways to address the ongoing challenge of card fraud and keep up with aggressive new fraud schemes, sophisticated solutions like Cooper Fraud Score will need to be a vital piece of the puzzle. Built on a modern technology stack and using a powerful continuous-learning predictive algorithmic approach to fraud prevention and detection, Cooper Fraud Score is the only unique supervised machine learning model focused on credit unions.

To learn more about how Cooper Fraud Score can help you prevent fraud before it happens and minimize losses, all while continuing to provide outstanding member service, reach out to your Co-op Fraud Prevention Consultant or contact us at 800.782.9042 or solutions@coop.org.

Fraud Prevention White Paper