This global marketplace was facing a difficult tradeoff: raise defenses and risk turning away real customers, or loosen controls and invite costly fraud. Pipl helped them break the tradeoff, improving approval rates, cutting chargebacks by more than two-thirds, and reducing operational burden in the process.
The existing system hadn’t been recalibrated in years. Rules weren’t evolving, and manual overrides were the only way to catch what the engine missed.
Risky behaviors such as mismatched names, freight forwarding, and identity stitching slipped through if they didn't trip a specific threshold.
The system could say "this looks risky," but never "this looks good." Without a trust tier, every gray-area order landed in review, even when risk signals were low.
Instead of rewriting rules, the team added scoring logic that mapped cleanly to the decision paths they already trusted. Pipl calibrated risk and trust thresholds using historical orders, then validated them against known outcomes. High-risk users got flagged earlier, high-trust users got approved faster, and review queues thinned out without sacrificing precision. It worked because it fit their system, not because it replaced it.
