SLA (Service Level Agreement)

How it works

SLAs define the minimum performance standard for a support operation. They protect customers by setting expectations and hold teams accountable with measurable targets. A typical SLA structure defines separate targets for first response time (FRT) and time to resolution (TTR), often with tiered urgency levels — a billing issue might have a 1-hour FRT SLA while a general product question has a 24-hour FRT SLA.

SLA management in a support tool involves three components: a timer that starts when the ticket arrives, a warning alert as the deadline approaches, and an escalation trigger when the deadline is breached. Effective SLA management also requires queue visibility — agents need to see which tickets are approaching breach so they can prioritize triage.

AI support changes SLA economics significantly. When the AI handles the first response, FRT SLAs become trivially easy to meet — the AI responds in seconds. The relevant SLA shifts to human-escalation response time: how fast does an agent pick up a conversation after the AI hands it off?

When you'll encounter it

SLAs appear in enterprise support contracts, support team performance reviews, and onboarding configurations for support platforms. You'll configure SLA windows during tool setup and monitor SLA performance in your support dashboard.