Reading about SLA policies in the abstract only gets you so far. Here is a worked example of building one from a real brief, so you can see how the pieces fit together before setting up your own.

The Brief

A support team wants two tiers of service: a strict tier for High Priority tickets, and a more relaxed tier for everything else. High Priority tickets should get a first response within one business hour and a resolution within eight business hours. Everything else should get a first response within four business hours and a resolution within two business days. Both tiers need an escalation path so a ticket that is about to breach does not simply sit unnoticed.

Step 1: Create the High Priority Policy

1. Go to Setup, then Automation, then SLAs, and create a new policy named something like "High Priority Response".
2. Set the criteria to Priority equals High, so only tickets flagged that way fall under this policy.
3. Set the first response target to 1 hour and the resolution target to 8 hours, both measured against business hours rather than the clock.

Step 2: Create the Standard Policy

1. Create a second policy named "Standard Response", with criteria set to Priority does not equal High so it catches everything else.
2. Set the first response target to 4 hours and the resolution target to 2 business days.

Step 3: Add a First Escalation Level to the High Priority Policy

1. From the High Priority Response policy, add an escalation set to trigger 30 minutes before the resolution target is due to breach.
2. Notify the assigned agent and their team lead, without reassigning the ticket at this stage, since a warning is usually enough at this level.

Step 4: Add a Second, Higher-Stakes Escalation

1. Add a second escalation to the same policy, set to trigger immediately if the resolution target is actually breached.
2. This time, reassign the ticket to a support manager rather than only sending a notification, so a genuinely overdue High Priority ticket gets direct ownership rather than a second warning nobody acts on.

What This Looks Like Day to Day

A High Priority ticket now shows agents a visible countdown timer from the moment it is created. Thirty minutes before the eight-hour resolution deadline, the agent and their lead get a heads-up. If that deadline is still missed, the ticket moves to a support manager automatically, with the original conversation history intact. Standard tickets follow the same pattern on a more relaxed timeline, without needing a separate manual process for anyone to remember to run.