Risk Matrix

Use this skill to run Risk Matrix as a practical consulting method, not as a generic framework explanation.

Method Notes

• Define risk as event + cause + consequence.
• Score likelihood, impact, and optional detectability, then assign owner and trigger.

Required Inputs

Collect or infer these inputs before execution:

• planned action or project
• risk list
• impact definitions
• owners

If an input is missing, do not block automatically. Mark it as missing, state the assumption used, and add a validation action.

When Not To Use

Do not use when numeric ratings would create false precision or when thresholds and owners are undefined. Treat legal, safety, and regulatory stop conditions separately from relative risk ranking.

Adjacent Methods

• pre-mortem: surface failure modes before a defined plan starts.
• scenario-planning: test strategy across uncertain external futures.

Step-by-Step Execution

Output Template

### 1. Scope And Scales
Decision / plan:
Horizon:
Likelihood scale:
Impact scale:

### 2. Risk Register
| Risk event | Cause | Consequence | Likelihood | Impact | Evidence |
|---|---|---|---|---|---|
|  |  |  |  |  |  |

### 3. Response
| Priority risk | Prevent / reduce / transfer / accept | Owner | Due |
|---|---|---|---|
|  |  |  |  |

### 4. Monitoring
| Risk | Leading signal | Threshold | Contingency |
|---|---|---|---|
|  |  |  |  |

### Evidence And Next Decision
- Confirmed facts:
- Assumptions:
- Missing evidence:
- Next action, owner, and timing:
- Expected signal and decision threshold:

Quality Gate

• Produce the method-specific outputs for Scope And Scales, Risk Register, Response; do not substitute a generic framework summary.
• Tie every material score, causal claim, or stakeholder judgment to evidence or label it as an assumption.
• Include at least one disconfirming check, sensitivity, alternative explanation, or failure condition appropriate to the method.
• Convert the result into a named action or decision with an owner or stakeholder, timing, and observable signal.
• State missing inputs and the smallest validation action instead of inventing precision.
• Keep wording professional and plain enough that a smart non-specialist can use the output directly.