🎯
Incident Severity Classifier
Determine incident severity levels and appropriate response procedures
How to use this classifier
- Answer questions about the incident's impact
- Review the automatically determined severity level
- Follow the response procedures for that severity
- Use the communication plan and escalation path provided
User Impact
Business Impact
Technical Factors
Operational Factors
Severity Classification
🟡
SEV3
Medium
Functionality degraded for subset of users. Should be addressed same day.
Response Time
Within 2 hours
Response Team
- •On-call Engineer
- •Team Lead
Communication Plan
- Notify on-call engineer
- Create ticket or incident thread
- Update team in daily standup
- Notify affected customers if applicable
- Daily status updates
Escalation Path
0-2 hours: On-call engineer
2-4 hours: Team Lead
4-8 hours: Engineering Manager
EOD: Review if not resolved
Immediate Actions
- →Begin investigation during business hours
- →Review recent deployments
- →Check monitoring and logs
- →Implement fix in next deployment
- →Add monitoring if gaps identified
- →Update documentation
ℹ️
Postmortem Optional
Document lessons learned in ticket
Severity Level Reference
🔴
SEV1
Critical
Examples:
• Complete service outage
• Data loss or corruption
• Security breach
• All users unable to access core features
• Payment processing completely down
🟠
SEV2
High
Examples:
• Major feature not working for many users
• Significant performance degradation
• Revenue-generating feature impaired
• Authentication issues affecting large percentage
• Critical API endpoints failing
🟡
SEV3
Medium
Examples:
• Feature degraded for subset of users
• Non-critical functionality impaired
• Workaround available but inconvenient
• Minor performance issues
• UI bugs affecting usability
⚪
SEV4
Low
Examples:
• Cosmetic UI issues
• Minor bugs with minimal impact
• Edge case failures
• Documentation errors
• Non-urgent improvements