How do I select a fraud detection solution for my business?

The rapid increase in digital use created a perfect storm for fraudsters to quickly find new ways to steal funds, capitalizing on consumers’ lack of familiarity with digital platforms and the resource constraints faced by many businesses.

In fact, from January 2020 to early January 2021, the Federal Trade Commission released that consumers reported over 275,000 complaints resulting in more than $210 million in COVID-19-related fraud loss. Because of this, it’s critical for businesses to anticipate potential new fraud schemes to prevent losses and protect customers.

To select a suitable fraud detection solution for your business, you need to think about a variety of factors. We’ve talked to several industry professionals to get their insight on the topic.

Shai Cohen, Sr. VP of Global Fraud Solutions, TransUnion

With the COVID-19 pandemic dramatically accelerating digital transformation, fraudsters have tried to take advantage and companies must adapt.

Businesses that come out on top will be those leveraging fraud prevention tools that not only provide great detection rates, but also friction-right experiences for consumers. For instance, 85% of global executives surveyed in an October 2020 global study by the Economist Intelligence Unit and TransUnion said they believe smooth digital transactions are “essential to business survival” rather than merely a competitive edge.

Here is what businesses should look for when selecting a friction-right digital fraud detection solution:

• Creating trust between a consumer and business by providing a comprehensive and concise view of the consumer
• Protecting consumer data and making the consumer experience personalized by having access to information about the consumer’s device, email, location, behavior patterns, etc.
• Being able to link a person’s online and offline data, and use a global data set to be able to accurately create a risk assessment of a consumer transaction using advanced analytics and machine learning capabilities. By doing so, a business is able to obtain a high fraud detection rate and low false positive rate.

Jimmy Fong, CCO, SEON

2021 brings a dizzying array of fraud detection options to online businesses. Here are 4 key considerations when selecting a partner to protect your business.

Choose your AI continuum. From strictly rules based to fully black box. The former gives you complete control but can be cumbersome and relies on a knowledgeable in house fraud team. The other end is perfect for extreme transaction volumes but offers little explanability. Fortunately, there is a middle ground with whitebox, supervised machine learning- you get the best of both worlds- granular rule based with machine learning making connections between disparate and complex data points.

Fully modular. Fraud detection technologies should fit your business, not the other way around. Fraudsters evolve and find ingenious workarounds to most point solutions. Modern fraud detection is a “net” approach where the latest cutting edge tools are used in combination to make it very, very hard for a fraudster to fool them all.

Rapid time to value. Results are very hard to predict. Try and select fraud technologies that allow you to test and show proof of value with no commitment, free trial periods.

Transparency. Modern, effective fraud tech should follow the best SAAS products where you see actual pricing, monthly contracts and free trials. Product value and risk should rest solely on the fraud detection partner.

Chris Halaschek, VP of Product, Pindrop

COVID-19 has forced individuals to rapidly adapt to handling all aspects of their life remotely, and contact centers have been one of the most significantly affected areas of global business – experiencing an unprecedented 42% increase year-over-year in call volumes (according to a study we conducted with Forrester Consulting).

Against this backdrop, CISO’s looking for fraud detection solutions should look for tools that can outsmart increasingly sophisticated fraudsters that are leveraging machine-to-machine vulnerabilities and tactics that combine IVR reconnaissance with phishing and vishing scams to gather data and perform account takeover (ATO) across channels.

Solutions that can monitor and measure risk across every call and account within contact centers and IVR environments, and ultimately predict and flag which accounts are most at-risk prior to an attack, are perfect for heightened threat landscapes.

A solution that is able to calculate account risk status and provide a score that indicates an active threat will help ease the job of security teams. Even better if these scores can be integrated through APIs to provide enterprises intelligence without disrupting existing systems and workflows seamlessly. These features will allow CISOs to improve overall organizational security posture by defending against the latest fraud tactics and providing a cross-channel view of potential fraud across the entire enterprise.

Yuval Marco, General Manager, Fraud & Authentication Management, NICE Actimize

Fraud is complex, dynamic, and unyielding. Advancements in technology, along with constantly changing fraud tactics, have created an evolving payments and fraud landscape. Further exacerbating the complexity of the fraud landscape is the proliferation of PII available on the dark web and the ever-increasing instances of data breaches.

We have also seen the growth and effectiveness of fraudsters utilizing stolen and synthetic ID’s. A recent Federal Reserve report stated that 85% to 95% of synthetic identities are not caught by the existing fraud model.

We believe that optimal fraud prevention strategies and implementations must be holistic and offer the ability to stream data in real-time to detect and defend against these emerging threats. Fraud prevention professionals should look for solutions that provide end-to-end coverage, using advanced technologies such as AI and Machine Learning to provide real-time dana processing to detect anomalous behaviors with the agility to adapt to shifting fraud.

While there are numerous tools and point solutions that address specific use cases, without this necessary holistic approach that connects and orchestrates your ecosystem of tools, fraudsters would still be able to exploit weaknesses created by the fragmentation of data and protection.

Andy Mortland, VP of Product, Accertify

With the increasing growth of online commerce, looming threats of data compromises, and pandemic-accelerated shifts in digital consumer behavior, fraud detection and mitigation are critical.

Here are five key fundamentals businesses should consider when selecting a partner:

• A platform with layered defenses – look for machine learning, rules, case management, device and user behavioral analytics. Using a platform with the flexibility to assess interactions with consumers from multiple touchpoints is key to defending against complex fraud schemes.
• Community – recognition of a ‘legitimate’ consumer through historical data, even if that consumer is new to you, can reduce inaccurate fraud alerts.
• People – seek a partner with a passion for solving complex challenges, and with experience in your field. Look for a good organizational fit, not just a vendor.
• Account focused – adopt a more holistic approach to account protection, for example, to defend against account opening and account takeover attacks.
• Technology – the solution you invest in should be dynamic in order to recognize new patterns and threats – it should continuously update, be built to scale, and adapt to your needs.

New fraud attacks and scams appear every day. By adopting a multi-faceted approach that incorporates the above principles, you can increase sales and conversion rates, provide a great customer experience, all while protecting your brand and reputation.

Michael Reitblat, CEO, Forter

The key is finding a solution that can maximize conversion rates while accurately identifying and blocking fraud. Today, many solutions falsely decline legitimate buyers due to a lack of user data.

Merchants should look for a solution that provides:

• Real-time decisions at every stage of the customer journey: many solutions still provide a risk score at the point of transaction requiring further manual investigation (opposed to an automated approve/deny decision). To protect the customer experience, real-time decisions must be made throughout the entire journey – from sign up, to login, to checkout, to returns. This defends against ATO, policy abuse, and additional attack vectors.
• Ability to maximize new user conversion rates: online transaction volumes for new users have increased more than 2x since the beginning of the global pandemic. Turning these new users into repeat buyers presents tremendous lifetime value for merchants. The challenge is new users are 5-7x more likely to get declined than repeat customers, and one false decline often ends the relationship.
• A robust global network: merchants relying on siloed data sources are unable to provide a comprehensive view of fraudulent and legitimate consumer behaviors. A larger data set beyond what the individual merchant has access to allows systems to accurately identify both. This allows merchants to approve as many transactions as possible and still block fraud.

Chris Ryan, Senior Fraud Solutions Business Consultant, Experian

It’s critical for businesses to consider a few things when selecting a fraud detection solution that can meet the demands of their customers while also mitigating risk.

First, it’s important that a business’s fraud risk problem doesn’t turn into a “what do we do now” problem. Identifying risk and addressing it go hand in hand, so the solution needs to trigger the next step automatically and in real time whenever possible. Also, the tool needs to apply the right type and amount of friction that reduces risk—in a way that makes customers feel comfortable and protected.

Companies should pursue providers who own and maintain their own data and use in-house experts to build advanced analytics. Data is important to understanding the difference between normal customer activity and out-of-character behavior. Expertise in using the data is indispensable. Robust analytics also help organizations identify different types of fraud, which is critical because fraud is dynamic. When fraud begins to rise, it’s important to zero in on the specific problem. One-size-fits-all solutions are not effective.

Finally, consider key performance indicators beyond fraud losses and incidents. Addressing fraud risk drives downstream costs. Friction from fraud prevention can push good customers away. Be sure to account for these impacts. They are often much larger than the fraud dollars you’re trying to save.