BigPay

The Challenge: Mule Activity Was Impacting Customer Experience

BigPay is one of Malaysia’s fastest-growing e-money platforms. Organized mule networks had identified BigPay as a target for moving stolen funds. BigPay was receiving over 1,000 reported mule cases per month which took a toll on internal processes and resources.

At one point, other financial institutions had begun declining legitimate BigPay top-up transactions that impacted customer experience. With Malaysia’s Fair Treatment of Financial Consumer Policy increasing accountability for customer protection, there was a stronger need to enhance fraud detection and response capabilities.

The Solution: Leveraging Cross-Channel Visibility for Rapid Threat Neutralization

Feedzai’s platform was already in place. The focus was translating patterns into detection logic. Feedzai’s Customer Success team worked directly with BigPay’s fraud analysts to build that capability, training on rule design, alert logic for suspicious inbound payments, and to decline rules based on funding velocity and beneficiary risk.

Once the team could build and deploy independently, they ran a structured pattern analysis. Mule accounts were predominantly opened in the same month they were used, and funds exited through rotating channels. The response:

• Cross-channel signal fusion: Rules evaluated the full payment journey, transfer and card activity together, to catch patterns such as third-party top-ups followed by crypto exits.
• Blocking cash-out channels: Every time the team identified a new exit route—crypto, ATMs, remittances, QR payments—rules were deployed quickly to reduce exposure and disrupt evolving laundering pathways.
• Simulation-first deployment: Every rule was validated before going live, compressing weeks of testing into hours.

Each closed exit produced cleaner data and a smaller case pool, making the next analysis sharper.

Two factors were decisive. First, cross-channel visibility: the ability to connect inbound funding events with outbound transaction behavior within a single analytical frame. Mule patterns only become legible when you can see the full journey. Second, speed of iteration. The ability to simulate a rule, validate it, deploy it, and observe results within hours, rather than weeks, meant faster response to emerging fraud patterns.

The compounding effect was equally important. As mule volumes fell, the remaining cases became analytically tractable. Smaller sample sizes made pattern identification easier, which drove further rule improvements, which drove volumes lower still.

The Results: Earlier Detection, Faster Response

BigPay’s fraud posture has been strengthened and is now evolving from reactive controls toward more predictive and adaptive prevention.

Catch Mules Earlier

With transaction monitoring stabilized, the next layer of defense brings signals upstream. Feedzai Digital Trust adds device intelligence, session behavior, and identity signals—flagging suspicious accounts before funds move.

II. Build resilience into the flow

Feedzai’s Rules Engine enables fraud teams to create detection logic from natural-language input, reducing the time from threat identification to live rule deployment. AI-assisted alert prioritization also helps analysts focus on the most critical cases which improves response speed and decision quality.

Due to this multi-layered detection approach and rapid threat response, BigPay neutralized a new mule network surge in Malaysia within 72 hours, reducing overall mule activity by 90%+ in 60 days.