Do I Need to Escalate?
A decision guide for knowing when to stop, raise risk, and involve an approver or response lead.
Use this page to decide when work must stop locally and move to a higher authority.
Escalation is not failure. It is the control that prevents unauthorized certainty.
1. Problem this page solves
Governed work fails when escalation is implicit, delayed, or personality-based.
This page defines when risk or uncertainty has moved beyond local decision authority.
2. What you should understand after reading
After this page, you should understand:
- which escalation triggers are in scope
- which risk or uncertainty patterns require escalation
- what to do before and after escalation
- when escalation is required even if execution still seems possible
3. Mechanism-first escalation path
- Identify the active task or decision. Clarify the current objective and the planned action.
- Identify escalation signals. Detect scope ambiguity, weak evidence, elevated risk, failed checks, or authority gaps.
- Check local resolution authority. Decide whether you are allowed to resolve the issue within current boundaries.
- If local resolution is not allowed, stop or hold. Do not continue high-risk or ambiguous execution.
- Escalate to the correct authority. Route the case to the approver or response lead with decision context.
- Preserve evidence and rationale. Make sure the escalation package is reviewable and actionable.
Escalation is required when the system cannot justify continued action within the current authority, evidence, or risk limits.
4. Observed vs inferred
| Layer | What is observed | What is inferred |
|---|---|---|
| Observed | scope conflicts, missing approvals, contradictory artifacts, failed checks, explicit risk signals | none beyond recorded signals |
| Inferred | judgment that uncertainty or risk is too high to continue safely | depends on operator interpretation and context pressure |
5. Trust assumptions
Escalation quality depends on assumptions that can fail:
- operators can underestimate risk
- evidence can look cleaner than the underlying reality
- time pressure can distort judgment
- escalation outcome depends on correct routing and preserved context
6. Explicit decision outcomes
| Outcome | Condition | Action |
|---|---|---|
| Continue locally | risk or uncertainty remains within authority and evidence limits | proceed under governed controls |
| Stop and escalate | authority, evidence, or risk boundary exceeded | halt local progression and escalate |
| Preserve and hold | escalation is needed but immediate execution is unsafe or unclear | preserve artifacts and wait for decision |
| Redirect to lab or non-production path | normal governed path is not applicable but a controlled exception is possible | use documented exception controls |
| Deny action pending clarification | requested action cannot be justified with current inputs | do not execute until clarified or approved |
7. Next-page handoff
Next, read Audiences for role-specific decision and escalation paths for operators, approvers, defenders, and integration authors.
| When to stop | Stop when the next action is intrusive, hard to reverse, or broader than the original task. |
| Escalation trigger | Escalate when scope, impact, authority, or evidence quality becomes uncertain. |
| Evidence required | Provide the objective, target, observations, actions already taken, current risks, and recommended next step. |
| Next path | What Evidence Is Required? |