Sample engagement · computed in your browser
AML — transaction monitoring scenarios
Five monitoring scenarios measured over 6,580 alerts and a reviewed below-the-line sample, against the institution's declared scenario inventory and its thresholds.
Mistakes we hid in this data on purpose
- — `dormant-account-reactivation` has never produced a SAR in nearly 400 alerts — a rule switched on years ago and never retired, because retiring one requires somebody to sign that it is safe to stop looking.
- — `wire-round-amount` fires more than any other scenario and converts at about half a per cent, and its alerts shift from retail to corporate halfway through the window.
- — `cross-border-high-risk` produces SARs BELOW its EUR 15,000 threshold, which is the finding a below-the-line exercise exists to produce: the threshold is set too high. `structuring-cash` has the opposite defect.
- — Roughly a quarter of alerts were never reviewed at all, and are carried as unreviewed rather than counted as cleared.
This is made-up data, and nothing on this page is precomputed — it runs through the real checking engine right now.
The evidence, and what was declared about it
- Book
- Alert log — synthetic, 6,580 alerts across five scenarios
- Source
- AlphaIQ synthetic alert log (aml-alert-log-dgp.json)
- Fixture
- aml-alert-log.csv
- Model
- Transaction Monitoring Scenario Suite v6
This sample has no pass/fail target on file — on purpose. Nobody handed us the institution’s own cutoffs, so each number below is reported but not graded, and that is exactly why the recommendation below reads the way it does.