Guide

How to set up SLAs in Jira Service Management

For Jira Service Management Cloud · Updated November 2026

Short answer: an SLA in Jira Service Management is a timer with three parts: when it starts, when it pauses, and when it stops. It also has one or more goals, such as "4 business hours for high priority". Set them up under Space settings → Request management → SLAs. Create a calendar for your working hours first, keep the number of SLAs small, and always pause while you're waiting for the customer.

Before you start: decide what you're promising

Most teams need only two SLAs:

SLA	Measures	Typical goals
Time to first response	How long until a customer hears from a person	Urgent 1h · High 4h · Normal 1 business day
Time to resolution	How long until the request is solved	Urgent 4h · High 1 business day · Normal 3–5 business days

Promise what your team can keep. An SLA that's breached on half the queue gets ignored, so start generous and tighten it once you have a month of data.

Step 1: Create a calendar for your working hours

Without a calendar, SLAs count 24/7, so a request that arrives on Friday evening breaches by Saturday morning. In the SLA settings, open Calendars and create one with:

Your time zone
Your working days and hours, such as Monday–Friday 09:00–18:00
Your public holidays. Add next year's before 1 January.

Teams that support customers 24/7 for urgent issues can use the 24/7 calendar on the urgent goal only.

Step 2: Create the SLA and set its conditions

Go to Space settings → Request management → SLAs and create a new SLA, or edit one of the defaults. Then set its conditions:

Start counting time when: usually Work item created. You can add several start conditions; the timer starts when any one of them happens.
Pause counting time when: almost always the Waiting for customer status. Otherwise you're measured on time the customer took to reply.
Finish counting time when:
- For first response, use Comment: For customers, so an internal note doesn't stop the clock.
- For resolution, use Resolution: Set.
- Several finish conditions also work as "any one of them".

Step 3: Add goals

Goals say how much time each kind of request gets. Each goal has a JQL filter, a target time and a calendar. Goals are checked from top to bottom, and the first match wins. Example for time to resolution:

Applies to (JQL)	Goal	Calendar
`priority = Highest`	4h	24/7
`priority = High`	8h	Business hours
`"Request Type" = "Get IT help"`	24h	Business hours
All remaining work items	40h	Business hours

Always keep the final catch-all goal. Without it, requests that match nothing have no SLA at all, and nobody notices.

Step 4: Show it to agents and act on it

Queues: add the SLA columns and sort by "time remaining", so agents work on what's about to breach first.
Automation: add a rule with the SLA threshold breached trigger, set to "will breach in 30 minutes", that notifies the assignee or the team channel.
Reports: the built-in SLA reports show success rates by week. Review them monthly, not daily.

Common mistakes

Mistake	What happens	Fix
No pause on "Waiting for customer"	You breach while the customer is on holiday	Add the pause condition
First response stops on any comment	An internal note "counts" as responding	Finish on Comment: For customers only
No working-hours calendar	Weekend requests breach before Monday	Add a business-hours calendar to each goal
Too many SLAs and goals	Nobody knows which one matters	Keep two SLAs and four or five goals
Duplicate requests from the same customer	One problem runs two or three SLA clocks and skews your reports	Merge duplicates so only one clock runs

Changing an SLA's conditions or goals can change the SLA values on existing requests. Make big changes at a quiet time, then check a few open requests afterwards.

Related guides

Want it done for you?

I set up SLAs, queues, calendars and alerts as part of a fixed-price Jira Service Management setup. See the setup service.