SLA targets require the Pro plan or higher.
The two targets
SLA targets are a single workspace-wide policy with two independent clocks:- First-response target — time from when a conversation starts to the first human reply. An AI reply does not stop this clock — only a reply from a teammate does, matching the “time to first human reply” convention used by Zendesk and Intercom.
- Resolution target — time from when a conversation starts to when it’s closed.
Configuring SLA targets
Go to Settings > SLA targets. This card is locked (with a Pro badge) on the Free plan.- Enable SLA targets — turns tracking on or off for the workspace.
- First-response target (minutes) — default 60.
- Resolution target (minutes) — default 1440 (24 hours).
- Count business hours only — when on, the SLA clock only advances during your configured business hours schedule; when off, targets are plain wall-clock time.
Business hours awareness
If Count business hours only is on, due dates are computed by walking forward through your weekly business hours schedule (configured in Settings > Business Hours) rather than counting continuous wall-clock minutes. A conversation that starts right before closing time won’t breach its first-response target overnight — the clock effectively pauses until you reopen. This uses the same business hours schedule and workspace timezone that gate Human Handoff availability. Snoozing a conversation also pauses both SLA clocks: the time spent snoozed is credited back to the due dates when it’s unsnoozed, so a long snooze never causes a false breach the moment the conversation wakes up. See Live Inbox for snooze behavior.What happens when a target is breached
A background check runs every 60 seconds across all workspaces with SLA enabled, comparing open conversations against their due dates. Each breach type (first-response, resolution) is flagged once per conversation and never re-flagged. In the Live Inbox, breached and at-risk conversations show a colored chip next to the conversation:- A red SLA breached badge once either target has been missed.
- An amber due in Xm countdown when a target is close to breaching (within 20% of its remaining window).
Notifications
Breaches also dispatch to your workspace’s connected notification channels — webhook, Slack, or Microsoft Teams — via the SLA Breach event toggle in Integrations. It’s on by default and fires for both first-response and resolution breaches, with the breach type included in the payload. There’s no email channel for this event; only webhook/Slack/Teams deliveries are supported.SLA in agent performance reporting
The Analytics page’s agent leaderboard includes an SLA column showing each agent’s combined hit rate across both targets, as a percentage with the number of eligible conversations in parentheses (for example,92% (48)). An em dash is shown when an agent has no SLA-eligible conversations yet.
A conversation only counts toward this metric once it’s closed. It counts as a “hit” when the relevant target existed, wasn’t flagged as breached, and — for the first-response target specifically — a human agent actually replied. It counts as a “miss” when the target existed but was flagged breached. Conversations without SLA due dates (SLA was disabled, or they predate enabling it) aren’t counted at all.
Availability by plan
See Billing & Plans for full plan details.
Tips
- Start with the defaults (60-minute first response, 24-hour resolution) and adjust once you have a baseline in the agent leaderboard.
- Turn on Count business hours only if your team doesn’t staff the inbox around the clock — otherwise every conversation opened overnight will breach before anyone is online.
- Watch the amber due in Xm chips in the inbox to catch conversations before they breach, not just after.