Transaction Monitoring
The continuous, rule- and analytics-driven surveillance of customer transactions to detect patterns consistent with money laundering, terrorist financing or fraud.
Definition
Transaction monitoring is the operational backbone of any AML programme: the continuous, automated review of customer transactions against the customer's declared profile and against typology-driven detection scenarios. Alerts feed an investigation queue that triggers further enquiry, EDD, customer-relationship action, and ultimately SAR filing.
Generations of TM systems
- Rule-based - deterministic scenarios; high false-positive rates.
- Risk-tiered rules - same engine, thresholds calibrated by customer-risk segment.
- Network analytics - links across customers, counterparties, devices.
- Supervised ML - alert triage, false-positive reduction.
- Unsupervised / graph ML - typology discovery, peer-group anomalies.
Common scenarios
- Velocity / volume above customer baseline.
- Round-amount transfers.
- Cross-border to high-risk jurisdictions.
- Counterparty risk (mixer, sanctioned wallet, mule cluster).
- Dormant-then-active accounts.
- Rapid in-out within minutes.
Regulatory anchor
FATF Recommendation 10 and Interpretive Note; EU AMLD/AMLR; EBA Guidelines on ML/TF risk factors; BaFin AuA; ACPR AML Guidelines. Supervisors increasingly expect documented model-validation cycles and explicit tuning histories.