> ## Documentation Index
> Fetch the complete documentation index at: https://docs.zenoo.com/llms.txt
> Use this file to discover all available pages before exploring further.
# How Do I Read Risk Scores?
> Understand the 4-dimension risk assessment in Zenoo — Customer, Geographic, Product/Service, and Transaction risk — and how to apply overrides.
# How Do I Read Risk Scores?
Every case in Zenoo has a risk assessment that scores the entity across four dimensions. Understanding these scores helps you prioritize your workload, apply the right level of due diligence, and justify your decisions to auditors.
## What you'll learn
* What the four risk dimensions are
* How scores and tiers are calculated
* How to read the risk assessment panel
* When and how to apply an override
* How risk scores change as you resolve alerts
## The four risk dimensions
Zenoo follows the FATF (Financial Action Task Force) methodology for risk assessment. Every entity is scored across four dimensions:
| Dimension | What it measures | Example risk factors |
| ------------------- | --------------------------------------- | ------------------------------------------------------------------------------------- |
| **Customer** | Who is this entity? | PEP status, sanctions matches, adverse media, complex ownership structures |
| **Geographic** | Where does this entity operate? | High-risk jurisdictions, FATF grey/black list countries, EU high-risk third countries |
| **Product/Service** | What products or services are involved? | Cash-intensive businesses, anonymous transactions, high-value products |
| **Transaction** | How does this entity transact? | Unusual patterns, cross-border flows, structuring, rapid movement of funds |
Each dimension gets:
* A **score** from 0 to 100
* A **tier**: High (70-100), Medium (40-69), or Low (0-39)
* A list of **risk factors** explaining what contributed to the score
## How is the overall risk calculated?
The overall risk tier uses a **highest-dimension-wins** approach:
| Scenario | Overall tier |
| --------------------------------------- | ------------ |
| Any dimension is High | **High** |
| No dimension is High, but any is Medium | **Medium** |
| All dimensions are Low | **Low** |
This is deliberately conservative. A single high-risk dimension (e.g., a PEP match pushing Customer risk to High) elevates the entire case to High risk, regardless of how low the other dimensions score.
## How do I read the risk panel?
The risk assessment panel appears on the case detail view. It shows:
At the top, a large badge shows the **overall tier** (High, Medium, or Low) with its color code:
* **High** — red badge
* **Medium** — amber badge
* **Low** — green badge
Next to it, the overall numeric score (0-100) provides more granularity.
Below the overall badge, four horizontal bars show each dimension:
* **Bar fill** — proportional to the score (0-100)
* **Tier label** — High, Medium, or Low
* **Score** — the numeric value
Click any dimension bar to expand its risk factors — the specific reasons driving that score.
Each dimension's expanded view shows a bulleted list of risk factors. For example, the Customer dimension might show:
* "PEP Match: Senior government official"
* "Adverse media: 3 articles related to fraud investigation"
* "Complex corporate structure with multiple layers"
These factors come from alert analysis, entity attributes, and country risk data.
Below the dimension breakdown, the panel shows recommended actions based on the risk tier:
* **High risk** — Enhanced Due Diligence required, senior reviewer approval needed, additional documentation required
* **Medium risk** — Standard review, consider additional checks
* **Low risk** — Standard review, routine processing
## How do I apply an override?
Sometimes you have context that the automated risk assessment does not capture. You can override the risk tier:
1. Click **Override** on the risk assessment panel
2. Select the new tier (High, Medium, or Low)
3. Write a **justification** explaining why you are overriding the calculated tier
4. Click **Apply Override**
The override is logged in the audit trail with your name, timestamp, and justification. The override tier takes precedence over the calculated tier for all downstream processing.
Overrides are visible to auditors. Always provide a clear, specific justification. "I know this person" is not sufficient. "Entity is a retired PEP with no current political connections, last held office 15 years ago" is.
## How do risk scores change during a case?
Risk scores are recalculated when:
* **An alert is resolved** — removing a high-severity alert can reduce the corresponding dimension score
* **A new assessment is created** — a fresh assessment using the current model rules replaces the previous one
* **An override is applied** — the override tier becomes the effective tier
As you resolve alerts on a case, you may see the risk score decrease as risk factors are cleared.
## What's next?
Alerts drive risk scores — resolving them changes the assessment.
Close the case once the risk tier is finalized.
AI research provides additional context for risk decisions.
Risk overrides are logged for regulatory review.