SLA Policies
SLA Concepts
Service Level Agreements (SLAs) in StackFlow define the maximum response and resolution times for ITSM records based on priority, category, and other criteria. StackFlow tracks three SLA metrics: Time to First Response (TTFR), Time to Resolution (TTR), and Uptime for service-level agreements.
- DynamoDB:
StackFlow_SLADefinitiontable with priority-to-response/resolution time mappings - DynamoDB:
StackFlow_SLAInstancetable tracking per-incident SLA timers - EventBridge: Rule
stackflow-sla-checkfiring every 5 minutes, targetingStackFlowSLACheckerLambda - SNS Topic:
stackflow-sla-alertswith email subscriptions for SLA breach notifications - Business Hours: At least one
StackFlow_BusinessHoursrecord for SLA pause logic
SLA timers start automatically when an incident is created and are paused/resumed based on state transitions (e.g., paused when status is "Pending Customer"). A breach occurs when the timer exceeds the configured threshold, triggering notifications to the SNS topic stackflow-breach-notifications.
Defining SLA Policies
{
"sla_policy": {
"name": "Standard Enterprise SLA",
"applies_to": {"record_type": "incident"},
"conditions": [
{
"field": "priority",
"operator": "equals",
"value": "P1",
"targets": {
"first_response_minutes": 15,
"resolution_hours": 4
}
},
{
"field": "priority",
"operator": "equals",
"value": "P2",
"targets": {
"first_response_minutes": 30,
"resolution_hours": 8
}
},
{
"field": "priority",
"operator": "equals",
"value": "P3",
"targets": {
"first_response_hours": 2,
"resolution_hours": 24
}
},
{
"field": "priority",
"operator": "equals",
"value": "P4",
"targets": {
"first_response_hours": 8,
"resolution_hours": 72
}
}
]
}
}
Business Hours
SLA timers can optionally exclude non-business hours. Configure business hour schedules in Workflows → Business Hours and link them to SLA policies. The business hours engine handles timezone conversions, organization-wide holidays, and location-specific exceptions.
SLA Metrics
| Metric | Description | Target (Example) |
|---|---|---|
| TTFR P1 | Time from creation to first agent response | ≤15 minutes |
| TTR P1 | Time from creation to resolution | ≤4 hours |
| TTFR P2 | Time from creation to first agent response | ≤30 minutes |
| TTR P2 | Time from creation to resolution | ≤8 hours |
| SLA Compliance Rate | % of records resolved within SLA | ≥95% |
Notifications & Escalation
SLA warning notifications fire at the configurable threshold (default 75% of time elapsed). SLA breach notifications fire immediately upon breach. Both notification types are routed through the SNS topic stackflow-sla-alerts, which can fan out to email, Slack, PagerDuty, or other endpoints via SNS subscriptions.