Help desk SLA examples you can copy today.

SLA basics in 60 seconds

A service level agreement (SLA) is a promise about speed and quality of support. In a help desk, it usually comes down to a few terms:

• First response time: from when the case is raised to the first human (or useful automated) reply. An auto-acknowledgement that says nothing doesn't count.
• Resolution time: from when the case is raised to when it's solved, or a workaround is in place and the customer agrees.
• Business hours vs calendar hours: say which. "4 hours" means something very different at 4:55 p.m. on a Friday.
• Pause conditions: the clock stops while you wait on the customer, and restarts when they reply.
• Breach: the moment a target is missed. What happens next is your escalation matrix.

Two related terms show up in larger teams. An OLA (operational level agreement) is an internal promise between teams, such as "Tier 2 picks up escalations within 30 minutes". An underpinning contract is the same promise from an outside vendor. A small business usually needs only the SLA and one or two OLAs.

Step 1: define P1 to P4 with real examples

SLAs fail when nobody agrees what "urgent" means. Define priority by impact (how many people are affected) and urgency (whether work or revenue is stopped), and give examples your team will recognize.

Step 2: help desk SLA examples (copyable tables)

Below are three sets of targets. Pick the one closest to how your business runs, then adjust. These are starting points we've seen work for service teams, not industry benchmarks. The best SLA is one your team hits 90% of the time or more.

Example A: small team, business hours only

Example B: clinic, spa or salon with front desk and back office

Example C: headquarters supporting locations or franchisees (internal SLA)

For franchise and multi-location teams, our internal help desk page adds a staff request form and a routing table you can pair with Example C.

Step 3: escalation matrix template

An escalation matrix says who takes over, when, and how they're told. It answers the question every agent asks at 4:50 p.m.: "Whose problem is this now?" The standard shape is three tiers: Tier 1 (front line), Tier 2 (specialist) and a manager. There are two kinds of escalation, and a good matrix covers both:

• Functional escalation: the case needs more skill or access, so it moves from Tier 1 to Tier 2.
• Hierarchical escalation: the case is at risk of breaching, or the customer is unhappy, so a manager steps in.

Copy the escalation matrix as text

ESCALATION MATRIX: [Business name], v1, [date]

Tier 1: [names / role]. Owns every new case. Solves P3 and P4.
  Escalate to Tier 2 when: P1 open 30 min | P2 open 4 h | needs access or a fix Tier 1 can't make.

Tier 2: [names / role]. Owns technical, billing and system cases.
  Escalate to Manager when: P1 at 50% of resolution target | any breach | customer asks for a manager | CSAT 1–2.

Manager: [name]. Owns breaches, complaints, refunds over $[limit].
  Escalate to Owner when: P1 past target | suspected data incident | key account at risk.

Owner: [name], [phone].

Out of hours: P1 only, call [on-call number]. Everything else waits for business hours.
Every escalation: reassign the case + add a 3-line note (what happened, what was tried, what's needed).

Step 4: an escalation process that works

1. Escalate the case, not the customer. The case moves with its history. The customer never repeats themselves.
2. Write a three-line handover note: what happened, what you tried, what you need.
3. Tell the customer. "I've passed this to our billing lead, who'll update you by 3 p.m." Name a time, not "soon".
4. The receiver acknowledges. Tier 2 picks up within its OLA, such as 30 minutes for P1.
5. De-escalate on resolution. The original agent closes the loop with the customer if they owned the relationship.
6. Review weekly. Count escalations by reason. If one reason repeats, write a macro or knowledge base article, or train Tier 1.

In JBMOnboard360, cases carry P1–P4 priorities with response and resolution targets, and move from Tier 1 to Tier 2 to a manager with the trail kept on the case. Because the case lives on the client record, Tier 2 sees the client's appointments, invoices and past cases without asking. See how that works in a CRM with help desk.

Step 5: SLA document template

Use this for a customer-facing SLA, or as the internal standard for your team. Keep it to one page.

SERVICE LEVEL AGREEMENT: [Business name] Support

1. Scope
   Covers support requests raised by [customers / locations / staff] through [channels].
   Excludes [e.g. custom work, third-party systems].

2. Support hours
   [Mon–Fri 9:00–17:00 ET]. P1 cover: [hours / on-call].
   Clocks run in business hours unless stated.

3. Priorities
   P1 Critical: [definition + example]
   P2 High:     [definition + example]
   P3 Normal:   [definition + example]
   P4 Low:      [definition + example]

4. Targets
   P1: first response [30 min], updates every [1 h], resolution [4 h]
   P2: first response [2 h], updates every [4 h], resolution [1 business day]
   P3: first response [1 business day], resolution [3 business days]
   P4: first response [2 business days], resolution [10 business days]

5. Clock rules
   Clock pauses while we wait for your reply and restarts when you respond.
   A case is resolved when [the fix is confirmed / no reply in 3 business days].

6. Escalation
   See the escalation matrix. You can ask for a manager at any time.

7. Reporting
   We report SLA attainment, first response time, resolution time and CSAT [monthly].

8. Review
   Targets reviewed every [quarter]. Owner: [name].

Step 6: macros that protect your SLA

Most breaches happen on the first response, because agents write every reply from scratch. These macros cover the moments that matter. Replace the brackets.

First response, P1

Hi [first name], thanks for flagging this. We've marked it urgent and [agent name] is working on it now. Next update by [time], even if it's not fixed yet.

First response, P3

Hi [first name], thanks for getting in touch. Your case number is [case #]. We'll come back to you by [day/time] with an answer or a next step.

Escalation to Tier 2

Hi [first name], I've passed this to [name], our [role], who can [what they can do]. They have the full history, so you won't need to repeat anything. Expect an update by [time].

Waiting on the customer (pauses the clock)

Hi [first name], to finish this we need [specific item]. Once we have it we'll [next step]. We'll keep the case open until [date].

Breach apology

Hi [first name], we promised you an answer by [time] and missed it. I'm sorry. Here's where it stands: [status]. [Manager name] is now overseeing it, and you'll hear from us by [time].

Resolution and CSAT

Hi [first name], [what we did]. This should now be sorted. If anything still isn't right, reply to this message and we'll pick it up again. You'll get a one-question survey; it helps us get better.

Step 7: help desk metrics and KPIs to track

Six numbers tell you whether your SLA is working. Review them weekly for the first month, then monthly.

Common SLA mistakes

• Copying an enterprise SLA. A 15-minute 24/7 P1 promise needs an on-call rota. If you don't have one, don't promise it.
• Counting auto-replies as a response. Customers don't.
• Letting agents set priority by mood. Use the definitions and examples above.
• No pause rule. Without it, cases waiting on the customer breach and the numbers lie.
• Never reviewing. Targets set once and forgotten drift away from reality.

Run these SLAs in JBMOnboard360

JBMOnboard360 Support CRM includes cases with P1–P4 SLAs, escalation from Tier 1 to Tier 2 to a manager, macros, a knowledge base the AI assistant answers from, and CSAT on every resolved case. Team members who raise cases are free. It's $115 per agent per month billed annually ($139 monthly), or included in Business at $399 per location with unlimited users. For clinics, a BAA comes with every plan; see the HIPAA-ready help desk. Still choosing a tool? Start with help desk for small business.