SLA and OLA are the backbone of any Service Desk operation. In this guide, we configure both in GLPI 11 from scratch, with real examples of response and resolution times by priority.
SLA vs OLA: understanding the difference
SLA (Service Level Agreement) is the agreement between IT and the client (internal or external). It defines what will be delivered and in how much time. Example: "High priority tickets will be responded to within 1 hour and resolved within 4 hours."
OLA (Operational Level Agreement) is the internal agreement between IT teams. It defines how teams will collaborate to meet the SLA. Example: "The infrastructure team must respond to escalations from the support team within 30 minutes."
Key concepts in GLPI
TTO – Time to Own (Response Time)
Maximum time between ticket opening and first assignment to a technician. It is the "response time" of the SLA.
TTR – Time to Resolve (Resolution Time)
Maximum time between ticket opening and solution. It is the "resolution time" of the SLA.
Configuring SLA in GLPI 11
1. Create SLA levels
Go to Configuration > SLAs and create an SLA for each priority:
- Very High SLA: TTO 15min, TTR 2h
- High SLA: TTO 30min, TTR 4h
- Medium SLA: TTO 1h, TTR 8h
- Low SLA: TTO 2h, TTR 24h
- Very Low SLA: TTO 4h, TTR 48h
2. Configure calendars
In Configuration > Calendars, define the service calendar (e.g., Mon-Fri 8am-6pm). SLA counts only business hours within the configured calendar.
3. Associate SLA with business rules
In Administration > Rules > Business rules for tickets, create rules that assign the correct SLA based on priority:
- Criterion: Priority = High
- Action: Assign SLA (TTO) = High SLA, Assign SLA (TTR) = High SLA
4. Configure escalation
In each SLA level, configure escalation actions:
- 75% of time: Notify the assigned technician
- 100% of time: Notify the group supervisor
- 150% of time: Reassign to the coordinator
Configuring OLA
OLAs follow the same logic, but control internal times between teams. In GLPI, configure them in Configuration > OLAs with the same TTO and TTR structure.
The typical scenario: when level 1 escalates to level 2, the level 2 OLA starts counting.
Best practices
- Start with few SLA levels (3-5) and adjust after 3 months of operation
- Use realistic calendars – don't configure 24x7 if your team only works business hours
- Configure alerts before violation (75%) to allow time for action
- Review SLAs quarterly based on real metrics
Next step
With SLAs configured, build your service catalog and define your Service Desk KPIs.
