Banking fraud platform pilot

Контекст

Розничный банк УЗ. Card fraud rising 25% YoY. Social engineering cases growing — customers переводят deньги мошенникам, public scandals. Existing fraud detection — fragmented per category (card processor, online banking, internal).

CRO задача — consolidated fraud loop.

Диагностика (4 недели)

Fraud loss assessment:

• Card fraud: 0.04% transaction volume (industry average 0.05).
• Social engineering: estimated $1.2M annualised losses, growing.
• Account takeover: rare detected, likely under-detected.
• Cross-category cases: invisible.

Подход (8 месяцев)

Месяцы 1-2. Architecture design. Customer-centric fraud platform consolidating signals across channels.

Месяцы 2-4. Phase 1 — card fraud upgrade. Real-time scoring engine, ML model retraining, false positive tuning.

Месяцы 4-6. Phase 2 — social engineering detection. Behavioural patterns + step-up authentication для suspicious transfers.

Месяцы 6-8. Phase 3 — cross-category investigation workspace. Fraud analysts работают с unified case context.

Результаты

После 8 месяцев:

• Card fraud loss reduction: 42%.
• Social engineering cases prevented: 89 cases, ~$680k saved.
• False positive rate -33% — better customer experience.
• Investigation throughput +60% per analyst.
• Customer NPS post-fraud-block (false positives): +14 points (faster resolution).

Что критически важно

Customer-centric approach — не «card fraud» vs «online fraud».

Step-up authentication критичен для social engineering.

Customer feedback loop в real-time — confirms / denies suspicion для model training.

Связанное

• /solutions/banking-fraud-platform/ — fraud platform
• /architecture/banking-realtime-decisioning/ — decisioning
• /use-cases/banking-suspicious-transaction-stepup/ — step-up
• /expertise/banking-fraud-discipline/ — fraud expertise