Fraud Prevention

How To Build A Fraud Team

There are different levels of fraud management an organization can implement, from completely in-house to completely outsourced. But for a company planning to build an internal team to manage and prevent fraud 24/7, it’s important to understand the value and limitations of this approach. In a recent whitepaper, Radial explores the evolution of a fraud investigation team to help businesses understand its true value.

According to Radial VP of Payments, Tax and Fraud, KC Fox, the cost of managing fraud in-house is often much higher than companies expect, and effective fraud management requires a mix of machine learning and human intelligence. The key, Fox said, is not just stopping fraud, but also enabling sales at the same time.

If your in-house team doesn’t know how to strike that balance, outsourcing will likely be the less expensive and more effective route. But if you do choose to take the DIY approach, here are some of the top things you need to know when building your own 24/7, professional fraud team.


  1. Fraud prevention is not the same as fraud management. “Everybody who has a fraud tool can stop fraud,” said Fox. But not everybody can strike the delicate balance that keeps most of the bad guys at bay and lets most of the good guys through. Fox sees it as a dial: Merchants can choose to do business only with customers they know and trust, thus preventing all fraud — but also losing business from potentially legitimate customers. Or, they can choose to allow all transactions, which reduces friction for the good customers — but also for the bad ones.
  2. Machine learning and artificial intelligence (AI) are not a substitute for human intelligence. Fraud tools for eCommerce often flag “risky” orders that are actually legitimate, and it takes human intervention to sort out which are which. Having an order wrongly flagged as “risky” disrupts the customer’s experience and may drive away their business, which harms company profits. Meanwhile, fraudsters know that fraud tools look for patterns, so they invent new ways to dupe the system, breaking pattern matches and getting through the defenses.
  3. The job of AI is to keep an eye on things when humans cannot. Repurposing the customer service team to moonlight as fraud investigators is better than doing nothing, but those people go home and sleep at night; they take the weekends off. What if a suspicious order comes in on Friday at 5:01 p.m.? The customer service team can’t address it until Monday morning, by which time the customer could have received the same order via one-day shipping from a competitor. AI never sleeps (and, Fox added, a professional risk management organization offers 24/7 monitoring).
  4. The job of manual analysts is NOT to find fraud that machines may have missed, but to win the transaction back. Say a customer is placing an order from overseas using a different credit card than usual and requesting overnight shipping to an address that does not match the billing address on file. A machine would peg that as fraud 100 percent of the time. Yet a human team could use data forensics and context (information gathered from sites and social networks like Facebook) to find that the cardholder is traveling abroad and trying to ship a gift to a relative.
  5. The best fraud and risk team includes human and non-human members. Let the machines do the heavy lifting. They will catch most of the bad guys and do so at hours when your in-house team isn’t watching. But when it comes to winning back transactions and customers, said Fox, that takes a human touch.


Building a professional fraud investigation team is a difficult and lengthy process. Even if a merchant succeeds in providing 24/7 dedicated coverage, it’s unlikely the results will be any better than the industry average due to the lack of expert data analytics backed by Big Data. That means a lot of money will still be left on the table. Knowing what it really takes to build a defense system from the ground up is essential in ensuring that you’re enabling both fraud prevention and sales.

To download: The Evolution of an Expert Fraud Investigation Team, fill out the form below:

First Name*

Last Name*



Work Email*


New PYMNTS Report: The CFO’s Guide To Digitizing B2B Payments – August 2020 

The CFO’s Guide To Digitizing B2B Payments, a PYMNTS and Comdata collaboration, examines how companies are updating their AP approaches to protect their cash flows, support their vendors and enable their financial departments to operate remotely.