When transaction-monitoring alerts overwhelm an analytics team, the first instinct is to adjust thresholds and scenarios. Often the cause sits earlier. The customer was rated at onboarding on incomplete or poorly reasoned information, and the monitoring system is now working from a weak baseline. This article explains why SAMA AML CTF customer risk profiling training deserves attention before the next tuning cycle, which four errors front-line staff most often repeat, and how role-play can correct them.
Why profiling errors surface as monitoring problems
Transaction monitoring compares behavior against an expectation. That expectation comes from the customer risk rating and the profile built at onboarding. If a cash-intensive business is rated low without a reasoned basis, its normal activity looks anomalous and generates alerts. If a genuinely high-risk customer is rated low, its activity may not alert at all. Both outcomes look like tuning problems, but neither is fixed by changing a threshold.
SAMA's AML/CTF rules, aligned with the FATF Recommendations, expect institutions to assess customer risk and apply measures in proportion to it. Whatever the exact policy wording, policy training tells staff what the rule says. It rarely shows them how a rating goes wrong in a real conversation.
Four errors the front line repeats
1. Rating cash-intensive businesses low without verifying the operational justification
Restaurants, retail chains, car dealers and similar businesses handle cash legitimately. The error is accepting that label as the justification. A rating should rest on whether the cash volumes make sense for the premises, the number of branches, the trading hours and the sector.
Training should teach relationship managers to ask for operational evidence and to record what they compared the declared activity against.
2. Accepting beneficial-ownership declarations without cross-referencing registry data
A customer's signed declaration is a starting point. Staff should compare it with commercial-registry data and the corporate documents provided. Mismatches in shareholders, managers or authorized signatories, and ownership layers that stop at another legal entity, should prompt further questions before a rating is assigned.
3. Missing geographic-risk elevation from correspondent relationships
Geographic risk is not only about where a customer is registered. A customer whose suppliers, buyers or banking partners sit in higher-risk jurisdictions carries exposure that a domestic address hides. Front-line staff often record the registration country and stop. Training should make them ask where money comes from and goes to, and which banks sit in between.
4. Failing to document the rating rationale in narrative form
A risk score without a reason is hard to use. Months later, a monitoring analyst reviewing an alert needs to know why the customer was rated as they were, what the expected activity was, and what was verified. A single dropdown selection tells them nothing. Short narrative notes written at onboarding save analyst time and make alert disposition more consistent.
A useful narrative covers four points:
- the nature of the business and the source of funds
- the expected transaction pattern and volumes
- what was verified, and against which source
- why the final rating was chosen
Three role-play scenarios
Role-play works because these errors occur in conversation, under commercial pressure. Each scenario below is built so that the relationship manager must ask a question, notice an inconsistency, and write the rationale.
Scenario 1: The restaurant chain requesting large cash deposits
A customer operating several restaurants asks to deposit substantial cash daily and wants a simple, low-friction onboarding. The facilitator, playing the customer, gives plausible but thin answers about branch revenue.
Participants practice:
- comparing declared cash volumes with the number of branches and the sector
- asking about card versus cash sales ratios
- recognizing deposits kept just under internal reporting levels as a possible structuring sign
- recording the expected deposit pattern so that later deviations are meaningful
Scenario 2: The trading company with a complex ownership structure
A trading firm presents a holding structure with several layers and a declaration naming one individual as beneficial owner. The registry extract, handed to participants, shows a different shareholder in one layer.
Participants practice:
- cross-referencing the declaration with registry data before rating
- asking why the structure is layered and what commercial purpose it serves
- deciding what must be resolved before onboarding proceeds and what to escalate
- writing a rationale that explains the unresolved points
Scenario 3: The logistics firm with cross-border payment patterns
A logistics company expects frequent payments to and from several countries, routed through intermediary banks. Its stated business is plausible, but the payment flow does not map neatly to its described services.
Participants practice:
- asking about counterparties, routes and the banks involved
- raising the geographic-risk rating when exposure sits in higher-risk jurisdictions
- spotting round-amount transfers, rapid in-and-out movements and payments with no clear link to freight activity as possible layering indicators
- documenting what was learned so analysts can interpret later activity
In each scenario, a legitimate and a suspicious version of the customer can be played with the same opening script. That teaches staff that the difference lies in the answers and the evidence, not in the sector label.
What a good outcome looks like
After effective training, onboarding files read differently. Ratings carry a stated reason. Ownership has been checked against the registry. Cash-intensive customers have an expected pattern on file. Analysts can open a file and understand the rating without calling the relationship manager. Alerts then reflect real deviations from a defined baseline, which is the point of doing this work before tuning.
Signs that it is time to act include repeated alerts on the same customer types, analysts who cannot find the basis for a rating, quality-assurance findings on missing ownership checks, and a tuning cycle approaching with no review of onboarding quality.
How Financial Crime Awareness Training is run
Our Financial Crime Awareness Training provides role-based AML/CTF, sanctions and fraud training for the board, front line, operations and compliance teams, with completion evidence for regulators. It is delivered in four steps:
- Profile: audiences, roles and risk exposure
- Tailor: cases and content for each audience
- Deliver: in-person or virtual sessions with assessments
- Evidence: records and an annual refresher plan
Scenarios like the three above are the kind of front-line red flags and case studies the program includes. Sessions run 1–3 days per audience. Deliverables include training materials in Arabic and English, attendance and assessment records, and an annual training plan. Participants who pass the assessment receive verifiable certificates, and content can be supplied in a format your learning platform supports. The program is aligned to SAMA AML/CTF, SAMA CFF and FATF.
Next step
If your monitoring alerts keep pointing back to onboarding files, start with the front line. Our Financial Crime Awareness Training can be tailored to your relationship managers' customer segments and run before your next tuning cycle. Contact CISO Consulting to discuss the audiences and scenarios that fit your institution.