AML Typology-to-Test-Case Mapping Worksheet
An AML examiner asks two questions: for each typology you say you monitor, can you fire the rule on a known case, and can you tune the rule without burning out analysts? Both depend on having a synthetic test corpus that exercises each typology with a known expected disposition. This worksheet maps each typology you monitor to the synthetic transactions and account patterns required, so the corpus and the rule book stay in sync.
What you walk away with
~22 min · 4 sections · 6 fields- A typology-by-typology test-case design — each row a synthetic case with a documented expected disposition.
- Identified gaps where a typology is on the monitoring list but lacks a representative test case.
- A cross-walk to the AML/BSA monitoring data checklist for each gap.
- An artifact compliance can hand to engineering for corpus construction.
Scope
Typology mapping
List the typologies the firm monitors and the synthetic case design that exercises each.
What the firm's monitoring system calls the rule.
Account configuration + transaction sequence that fires the rule. Be specific: amounts, channels, counter-parties, timing.
What the system should produce — alert level, escalation path, SAR conversion expectation.
What the firm's monitoring system calls the rule.
Account configuration + transaction sequence that fires the rule. Be specific: amounts, channels, counter-parties, timing.
What the system should produce — alert level, escalation path, SAR conversion expectation.
What the firm's monitoring system calls the rule.
Account configuration + transaction sequence that fires the rule. Be specific: amounts, channels, counter-parties, timing.
What the system should produce — alert level, escalation path, SAR conversion expectation.
What the firm's monitoring system calls the rule.
Account configuration + transaction sequence that fires the rule. Be specific: amounts, channels, counter-parties, timing.
What the system should produce — alert level, escalation path, SAR conversion expectation.
Tuning posture
How does the firm tune thresholds against false-positive volume?
Approximate FP% across all alerts in the last 90 days.
Coverage roll-up
Rough indicator only — for a granular score, run the AML Program Data-Coverage Scorecard.
Lower is better. Combines under-coverage of typologies with tuning weakness.
Close the gaps
Each row with status 'gap' is an archetype to add. The AML/BSA monitoring data checklist names the data signals required to fire each rule; the AML Program Data-Coverage Scorecard scores the firm's overall posture.
Key takeaways
- A monitoring rule without a documented synthetic test case is a rule the firm cannot defend in an exam.
- Tuning against production data alone is structurally lagging — synthetic corpora let the firm test threshold changes before deploying.
- Elder financial exploitation is the most under-monitored typology in wealth-tech firms. Worth its own row.
- PEP and sanctions screening cadence is rarely the bottleneck; it's the adjacency screening (family, controlled entities) that produces findings.
FAQ
We outsource AML monitoring to a vendor. Does this still apply?
Yes. The firm is responsible regardless of vendor. Use this worksheet to document what the vendor's rules are designed to fire, what they don't fire, and the synthetic cases that prove it. If the vendor can't help you populate this, that's a vendor-management finding.
How granular should the rule list be?
One row per named rule the monitoring system runs. If the system runs three structuring rules with different thresholds, that's three rows.