How to Design Service Desk SLAs That Actually Hold Up
An SLA is a promise about speed, written down. Teams skip them because they sound like enterprise process, but without one every request competes on who complains loudest rather than what is actually urgent.
Two clocks, not one
- First response — how long before a human acknowledges. This is what stops people re-sending and escalating.
- Resolution — how long before the service is restored.
Track both. A team that resolves quickly but never acknowledges still feels unresponsive; one that replies instantly and then goes quiet for a week creates a different kind of distrust.
Derive priority, do not let people pick it
If requesters choose their own priority, everything becomes urgent. Derive it from two answers instead: impact (how many people are affected) and urgency (how time-sensitive it is). One person unable to print is low impact, moderate urgency. A site-wide mail outage is high on both. The matrix produces the priority, and the priority selects the SLA.
Starting targets
- P1 — critical: respond in 15–30 minutes, resolve same day.
- P2 — high: respond within 2 hours, resolve in 1 business day.
- P3 — normal: respond in 1 business day, resolve in 3.
- P4 — low: best effort, no hard promise.
Adjust to your team size and coverage. The failure mode is promising numbers you miss consistently — a target you breach every week teaches people the SLA means nothing.
Count only the hours you actually work
If your desk runs 9 to 6, a ticket arriving at 8pm should not burn eleven hours of response time overnight. SLAs must run against a business-hours calendar, and pause while a ticket is legitimately waiting on the requester rather than on you. Without both, your dashboard is permanently red and everyone stops looking at it.
Act before the breach, not after
The value is in what happens at 80% of the clock, not in the breach report. A rule that escalates or reassigns before a target is missed prevents the failure; a report that lists last month's breaches only documents it.
SupportCentral derives priority from impact and urgency, runs SLA timers against your business hours, pauses them while you are waiting on the requester, and can escalate before a target is missed. See which record type carries an SLA in the first place.