Inside Enterprise Strategies for Fighting First Party Fraud at Scale – with Leaders from Justt and Walmart

This article is sponsored by Justt and was written, edited, and published in alignment with our Emerj sponsored content guidelines. Learn more about our thought leadership and content creation services on our Emerj Media Services page. 

Fraud and chargebacks present a massive and growing financial burden for merchants and financial institutions. According to Visa’s Fall 2024 Biannual Threats Report, fraud schemes are increasingly sophisticated, with threat actors leveraging system misconfigurations and vulnerabilities to exploit both merchants and cardholders. 

For example, Purchase Return Authorization (PRA) attacks, where fraudsters compromise merchant gateways to initiate refunds for nonexistent purchases, have surged by 81% in the first half of 2024 compared to the previous six months, with each successful attack costing financial institutions an average of $184,000.

Global chargeback volume is expected to grow 24% from 2025 to 2028, reaching 324 million transactions annually, according to Mastercard’s 2025 state of chargebacks report, based on research from Datos Insights. 

Mastercard also reports that, on average, financial institutions in the U.S. must hire one back-office full-time employee for every $13,000-$14,000 of cardholder disputes annually. For most financial institutions in the U.S., this translates to more than 200 full-time equivalents (FTEs), representing millions of dollars in personnel costs annually.

Each disputed transaction costs financial institutions $9.08 to $10.32 to process. Multiplied by an estimated 261 million chargebacks generated annually in 2025, this represents trillions of dollars of total expenditures across all financial institutions each year.

Most organizations still rely on legacy rule-based systems and manual chargeback processes that fail to keep up with modern fraud tactics.

According to research from the European Centre for Research Training and Development, UK, advanced analytics can analyze large volumes of data and identify subtle patterns and anomalies indicative of fraud that may go unnoticed by human analysts or rule-based systems. Deeper insights from advanced analytics, as shown in the study, are expected to lead to higher detection accuracy and lower false positive rates.

Emerj Senior Editor Matthew DeMello recently sat down with Roenen Ben-Ami, Co-founder and Chief Risk Officer at Justt, and Naveen Kumar, Director of Financial Crimes at Walmart, to discuss how fraud and chargeback challenges are evolving and how businesses can respond with more innovative data-driven strategies.

Their conversation highlights the growing complexity of fraud and dispute management, underscoring the need for integrated data, behavioral insights, automation, and cross-functional collaboration to enhance detection, mitigate losses, and safeguard the customer experience.

The following subsections examine key insights from these conversations listed below:

• Automating chargeback recovery and enhancing win rates: Improving chargeback win rates by automating case-specific responses using integrated PSP, merchant, and third-party data through a method called dynamic arguments.

• Expanding fraud detection and aligning risk management: Broadening fraud detection beyond transactions to include behavioral patterns and disputes while coordinating fraud and risk teams to balance prevention with customer experience and regulatory demands.

Automating Chargeback Recovery and Enhancing Win Rates

Episode: Tackling Chargeback Challenges at Scale with Machine Learning and Dynamic Arguments – with Roenen Ben-Ami at Justt

Guest: Roenen Ben-Ami, Co-founder and Chief Risk Officer, Justt

Expertise: Leadership, Business Analytics

Brief Recognition: Prior to helping found Justt, he built the chargeback and merchant risk teams that successfully recover millions of dollars annually at the payments service provider Simplex. Before that role, he served in an elite military intelligence unit in the Israel Defense Forces for nine years, attaining the rank of captain. 

Roenen opens the conversation on the ‘AI in Business’ podcast by explaining how chargebacks work in the retail industry. He says customers who dispute a transaction go to their issuing bank instead of the merchant to reclaim their money. The bank processes the chargeback, and the funds are returned to the customer. 

However, these funds are taken directly from the merchant, who – by the default, adjudication of the process – is basically treated as guilty until proven innocent. In other words, the system favors the customer by default, placing the burden of proof on the merchant to demonstrate that the chargeback is illegitimate, while the transaction behind the chargeback is legitimate.

He further explains that while merchants can dispute chargebacks by providing documentation, the process is complex and varies based on the reason code assigned to the chargeback. Typically, the issuing bank reviews the documentation. However, in some cases, it may escalate to acquiring banks or card networks, such as Visa and MasterCard.  

Chargebacks can stem from: 

• Fraud claims: Where the merchant must prove the actual cardholder made the purchase
• Service-related disputes: Where the merchant must show goods or services were correctly delivered). 

Roenen then mentions that his company, Justt, has developed a smarter and more scalable approach to handling chargebacks by leveraging data and automation effectively. The first step in this process is integrating data from multiple sources, including payment processors, merchant data, and third-party enrichment data. 

Payment processors such as Stripe, Adyen, and Worldpay provide essential transaction data, but some operate on older systems that require manual extraction.  Additionally, third-party enrichment data, such as IP addresses, phone numbers, and billing information, helps establish connections to determine whether a transaction is legitimate or fraudulent.  

Instead of relying on static templates, Justt’s system uses a method referred to as dynamic arguments (sometimes referred to as dynamic parameters elsewhere in data science), which ensures that the evidence presented is tailored to the specific situation. The placement of supporting arguments and screenshots is also adjusted based on what is most effective for winning a case.

By customizing responses this way, Roenen tells Emerj’s executive podcast audience that Justt increases the likelihood of a favorable outcome for merchants:

“Behind the scenes, our data science, R&D, and domain expert teams collaboratively define the foundation of Justt’s platform — establishing dynamic arguments, determining their adaptability, and codifying the logic for their deployment.

Once a customer is onboarded, the system operates at full scale, automatically managing chargeback volumes ranging from hundreds to hundreds of thousands. By selecting the optimal arguments for each individual case, the platform delivers tailored responses with greater precision than manual processes, reducing human error and significantly improving recovery outcomes.”

— Roenen Ben-Ami, Co-founder and Chief Risk Officer, Justt

He also shares with Emerj that Justt improves merchants’ chargeback win rates by leveraging three key data sources: 

• PSP data
• Third-party data, and 
• Additional merchant data.

Once a higher win rate is achieved, Justt further optimizes results by analyzing subcategories within the win rate, such as specific issuers, reason codes, and card schemes. 

Once a merchant adopts Justt’s solution, the system primarily operates independently, with minimal involvement required from the merchant. However, providing additional merchant data can further improve the win rate. 

If merchants want to enhance results, Roenen says that they can provide additional data in multiple ways: 

• Submitting a CSV file weekly, which only takes 20 minutes or 
• Integrating with Justt’s API for an entirely hands-off process. 

The latter enables the system to ingest the additional data and improve chargeback outcomes.

Expanding Fraud Detection and Aligning Risk Management

Episode: Evolving Expectations in Fraud and Risk Response from Dispute to Detection with Naveen Kumar from Walmart

Guest: Naveen Kumar, Director-Financial Crimes, Walmart

Expertise: Regulatory Compliance, Fraud and Threat Detection

Brief Recognition: Naveen has over 16 years of experience in AML, Insider Risk, Fraud, and Sanctions. Previously, he has worked with PwC and Stellaris Health Network. He holds a Master’s in Science in data modeling from The Rochester Institute of Technology. 

Naveen, in his podcast, discusses how the definition of fraud is rapidly expanding, moving beyond traditional notions such as stolen identities or unauthorized transactions. Today, it also includes more complex and harder-to-detect forms such as app misuse, social engineering, synthetic identities, and policy exploitation.

He emphasizes that fraud often doesn’t appear as fraud at first glance; what may seem like a simple customer dispute could be first-party fraud.

Naveen goes on to point out that fraud teams today are expected to move beyond simply reviewing transactions and instead focus on holistic behavioral analysis. It involves understanding user intent and identifying patterns of behavior rather than just focusing on outcomes. 

He also emphasizes the importance of cross-channel fraud prevention, encompassing digital, phone, and in-person interactions, as fraud is no longer confined to a single channel.

In addition to these operational shifts, he notes that there is a growing regulatory expectation for institutions to quantify their exposure to fraud. These efforts are leading to a push toward a risk-based approach that prioritizes prevention over control. 

As a result, fraud and risk teams are increasingly collaborating, moving beyond traditional rule-based detection. Naveen describes how most teams are now incorporating a combination of supervised and unsupervised machine learning algorithms to create more comprehensive fraud detection systems.

Naveen then reflects on how the fast pace of operations can cause teams to overlook essential questions, such as whether an incident is the result of fraud, abuse, or a system loophole. He explains that this is where evolving expectations come into play.

Organizations are beginning to recognize the need to separate dispute management from fraud detection, as a dispute can often serve as a valuable data point or early warning sign of a larger issue:

“When organizations treat disputes primarily as a customer experience issue, they risk overlooking early indicators of fraud. Disputes and chargebacks carry significant financial impact — so it’s critical to distinguish between first-party and third-party fraud. The key lies in leveraging early risk signals within the data to proactively identify and address underlying issues.”

— Naveen Kumar, Director of Financial Crimes at Walmart

He also explains that it’s essential to consider what the actual goal is and what the cost is. No matter how effective fraud detection models are, there will always be false positives, which incur business costs and customer friction. 

Naveen punctuates his point here by reminding the executive audience that fraud prevention efforts must align with the business, ensuring that when operations inevitably go wrong, the organization’s reputation and regulatory compliance are protected. He emphasizes balancing fraud prevention with minimizing customer friction and managing regulatory risks.