Start from a working process, not a blank canvas. Our team fits it to each entity, on the tools you already run. Customers see 45% fewer licenses and 96% fewer execution issues.
A backlog of reviews worked oldest-first by a team that also handles escalations.
What changes
Reviews open when something changes, not only when they fall due. Analysts spend their time on judgment, not collection.
Works with
Fenergo
Pega
Your screening engine
Your risk model
Routed by risk, not by queue
The checks and who decides follow the customer's risk rating. A review also opens when something changes, such as a new director, an alert or adverse media.
Low riskAt least every 5 years
What runs
Company and ownership data refreshed from registers
Alerts since the last review checked
Compared with the last review
Who decides
No material change: an agent prepares the comparison, an analyst confirms it in one step, and a sample goes to quality control.
SLA 5 days
Medium riskAt least every 3 years
What runs
Documents collected from the customer
Alerts since the last review checked
Risk rated by your model, the file prepared
Who decides
An analyst decides on the prepared file.
SLA 2 days with the analyst
High riskAt least every year
What runs
Documents and source of wealth corroborated
Screening and adverse media
Enhanced due diligence drafted
Who decides
A senior analyst decides and a second confirms. Keeping a politically exposed person needs senior management approval, and an exit needs compliance sign-off.
SLA 5 days with the analysts
The file an analyst opens
By the time a review reaches a person, the documents are requested and the alerts are triaged. The analyst's time goes to judgment, not collection.
C-8815Opened on a register change · Medium risk · Opened 2 days ago
Why it opened
New director registered this week, from the company register
Transaction volume above the expected range for six weeks
Screening
1 possible sanctions match on the new director, from your screening engine
The engine flags a different date of birth and nationality. Proposed as a false positive, for the analyst to confirm.
Documents
4 of 5 on file
Proof of address requested from the customer, reminder in 7 days
Risk rating
Medium, unchanged, from your risk model
Drivers: country low, product medium, activity rising
Confirm refreshRequest informationEscalate to enhanced review
Confirm refresh opens once the match is cleared and the proof of address is in.
An illustrative case. The analyst's decision is recorded with its reason.
Inside each stage
Choose a stage to see who does each step and what it leaves on the case.
Review opens
Day 0
Steps
TriggerDue date reachedFrom the review cycle your policy sets per risk tier
APIOr an event arrivesA register change, a screening alert or adverse media
APIRouted by risk tierLow, medium or high, from the current rating
Left on the case: why the review opened, and when.
Systems it touches
Client lifecycle management, such as Fenergo or Pega
Company registers
Your screening engine
Controls
Every customer has a next review date set by policy
An event opens a review even when none is due
When it doesn't fit
An event for a customer whose review is already open joins that review instead of opening a second one.
Collect documents
Days 1 to 30
Steps
APIRPARegisters checkedCompany, ownership and director data by API, RPA where a register has none
APIDocuments requestedOnly what the registers cannot show, through your portal or email
APIPeopleReminders sentAt 7 and 14 days, then the relationship manager is asked
AI agentsDocuments checkedThe document agent compares them with the register data and flags gaps
Left on the case: each document, where it came from and when it arrived.
Systems it touches
Company registers
Customer portal or email
Document store
Controls
The customer is asked only for what the registers cannot provide
Each document is linked to the requirement it meets
When it doesn't fit
A customer who does not reply after the last reminder goes to the relationship manager, then to the exit route your policy sets.
Screening alerts
Day 1 · SLA 1h
Steps
APIScreening runCustomer, directors and owners, by your screening engine
AI agentsAlerts comparedThe screening agent compares each alert with the file and proposes a disposition
AI agentsAdverse media checkedThe media agent records each finding with its source and date
Left on the case: each alert, the data the engine matched on and the proposed disposition.
Systems it touches
Your screening engine
Adverse media sources
Controls
No alert is closed without an analyst
When it doesn't fit
A likely true match stops the review and goes straight to an analyst, with what the screening agent compared.
Risk rating
Day 1 · SLA 2h
Steps
APIRating runBy your risk model, on the refreshed data
AI agentsDrivers explainedThe explainer agent writes why the rating moved, or why it did not
APITier checkedA higher rating moves the review to that tier's checks and approvers
Left on the case: the rating, its inputs and the agent's explanation.
Systems it touches
Your customer risk model
Client lifecycle management
Controls
The agent never sets or overrides the rating
A rating change re-routes the review
When it doesn't fit
An analyst who disagrees with the model can override the rating, with a reason and a second approver.
Analyst decision
After triage · SLA by risk tier
Steps
PeopleFile reviewedDocuments, alerts, rating and the comparison in one place
PeopleDecision recordedConfirm, request information or escalate, with a reason
PeopleSecond approvalFor high risk, and senior management for a politically exposed person
Left on the case: the decision, its reason, the analyst and any second approver.
Systems it touches
Client lifecycle management, such as Fenergo or Pega
Controls
An analyst confirms every review
Four eyes for high risk, senior management for a politically exposed person, compliance for an exit
When it doesn't fit
Escalation moves the review to enhanced due diligence, with its own approvers and SLA.
Record updated
On decision
Steps
APIProfile updatedRating, documents and the next review date, by API
APIRPACore banking updatedRisk flags by API, RPA where a system has no interface
CloseCase closedEvery check and decision on one record
Left on the case: the updated records' references and the next review date.
Systems it touches
Client lifecycle management, such as Fenergo or Pega
Core banking
Document store
Controls
The next review date follows the new rating
When it doesn't fit
An update that fails is retried only once the target shows it did not apply, then goes to operations with the error.
What your reviewers will ask
IT security, compliance, model risk and audit each sign off before go-live. Their first questions, answered.
IT security
Where does customer data go?
Nowhere. Optomo runs on-premise, in your network, so customer documents, screening alerts and agent reasoning stay in it. Optomo is ISO 27001 certified.
Compliance
Does an AI decide on a customer?
No. An analyst confirms every review. High risk needs a second analyst, keeping a politically exposed person needs senior management approval, and an exit needs compliance sign-off.
Model risk
Whose model rates the risk?
Yours. An agent explains what drove the rating but never sets or overrides it. The rating, its inputs and the explanation are recorded on the case.
Audit
Can we show a regulator why?
Each decision is recorded with its reason, the analyst and the evidence on file. Enterprise contracts give your auditor direct access to the controls Optomo runs.