Run: Tier 3 Engineering Escalation SOP

The Tier 1 or Tier 2 owner reviews the issue against escalation criteria, including repeated failure, unresolved root cause, customer impact, safety risk, data integrity risk, or need for code-level or architecture-level analysis. The owner records the reason for escalation and the current severity, priority, and business impact.

The assigned owner performs the approved remediation steps, documents the fix, and verifies the result with the requester or monitoring data. If the issue remains unresolved after the attempt, the owner returns to the escalation decision and prepares a Tier 3 handoff.

The current owner documents the incident summary, affected service or component, timestamps, symptoms, customer or operational impact, severity, priority, recent changes, attempted troubleshooting, logs, screenshots, error codes, and any known workarounds. The owner assigns the correct Tier 3 queue or engineer and includes all relevant ticket links.

The current owner posts the escalation in the designated communication channel, tags the Tier 3 role or on-call engineer, and confirms the receiving owner. The handoff includes urgency, response expectation, and any immediate containment actions already taken.

The Tier 3 engineer evaluates whether to apply a rollback, disable a feature flag, isolate a failing component, or implement another approved containment action. The engineer records the action taken, the rationale, and the observed effect on service behavior.

The Tier 3 engineer analyzes logs, metrics, traces, configuration history, code changes, dependency behavior, and reproduction steps. The engineer distinguishes between symptom, contributing factor, and root cause, and records whether the issue is a defect, configuration error, environment issue, or process gap.

The Tier 3 engineer tests the proposed fix, confirms the service returns to normal tolerance, and verifies that the issue no longer reproduces. If the fix cannot be validated, the engineer records the deviation and reopens investigation.

The engineer records the final status, root cause summary, corrective action, verification result, residual risk, and any follow-up tasks. The record should be complete enough to satisfy documented information requirements and support future audits or trend analysis.

The Tier 3 engineer identifies whether the issue requires a bug fix, design change, monitoring improvement, documentation update, or training update. The engineer creates or links the appropriate product feedback item and assigns it to the product or engineering owner with clear acceptance criteria.

The owner sends a closure update to the requester, incident manager, and relevant stakeholders, including the final resolution, any workaround, and any follow-up actions. The ticket is then closed only after all required documentation and feedback links are complete.