Usage Scope

Product overview

First-party fraud (FPF) occurs when an individual uses their own real identity to open accounts, access credit, or transact—then commits a dishonest act such as:

• Disputing legitimate transactions (chargeback abuse)
• Defaulting on loans or payments
• Exploiting refund or return policies
• Securing funds with no intent to repay

Unlike third-party or synthetic fraud, there are no stolen credentials or fake identities. This makes FPF uniquely hard to detect with traditional fraud tools. The cost is enormous: U.S. businesses face $100B+ in annual losses, with financial services, fintechs, BNPL, telcos, gaming, and e-commerce among the hardest hit.

Industries it fits

• Financial services: Detect repeat fraudsters at account opening or during payments
• Lending & BNPL: Spot borrowers likely to commit chargeback abuse or defaults
• Gaming: Prevent duplicate accounts and fraudulent sign-ups
• Government & benefits: Block multiple claims tied to the same identity
• E-commerce & subscriptions: Flag consumers with dispute-heavy histories before renewals
• Payments & wallets: Protect against fraudulent ACH, card, P2P, and wire transactions

Business and operational outcomes

1. For compliance and risk teams

• Reduce fraud losses tied to chargebacks, disputes, and defaults
• Identify high-risk applicants with prior confirmed fraudulent activity
• Detailed reason codes support transparent decisioning and audits
• Coverage expansion through consortium contributions

2. For growth and onboarding teams

• Safer onboarding: Catch high-risk applicants before they enter your ecosystem
• Protect portfolio health: Block consumers with histories of payment abuse
• Configurable thresholds: Tune risk signals and fraud scores to align with risk appetite
• Support for multiple flows: From new applications to mid-journey payment verification

3. For support and operations

• Fewer escalations: Reduce dispute handling by blocking known abusers upfront
• Clear decision signals: Agents see risk signals, reason codes, and fraud scores
• Operational efficiency: Automated fraud checks reduce manual reviews

Performance considerations

• Latency: Typical response time for RiskOS API is under 500ms per call.

• Accuracy: At 10% risk population, Sigma FPF captures >50% of FPF with high precision; at 3% risk depth, it can flag 30%+ of true first-party fraud.

• Scalability: Supports enterprise volume and big data workloads.

Known Limitations

• Primary coverage is US; international signals are planned but not yet fully supported
• Not a consumer reporting agency: cannot be used for FCRA-regulated adverse action
• Dependent on ongoing, high-quality consortium data contributions for optimal performance.
• Rate limits and provisioning controls apply for very high-volume or bulk/batch use cases