Service Level Agreement (SLA) Template: What It Is and What to Include

A Service Level Agreement is a contract between a service provider and a customer that defines exactly what level of service will be delivered, how it will be measured, and what happens when it falls short. Without one, "we'll keep your systems running" means nothing. With one, it means "99.9% uptime, 4-hour response time, $500 credit per incident."

SLAs are standard in IT services, cloud hosting, managed services, and any professional service relationship with ongoing delivery obligations.

The three types of SLA

Service-based SLA: Covers all customers receiving a specific service. Example: A cloud provider's standard uptime SLA applies to all customers on the same infrastructure tier.

Customer-based SLA: Customized for a specific customer, often for enterprise or high-value accounts. Negotiated SLAs with larger guarantees, dedicated support, or custom escalation paths.

Multi-level SLA: A combination — defines corporate-level baselines, service-specific additions, and customer-specific terms layered on top.

What every SLA must include

1. Service Description

Define exactly what service is being provided. Be specific. "IT support" is not a service description. "24/7 monitoring of the client's production infrastructure across AWS us-east-1 and us-west-2, with incident response for P1 and P2 severity events" is.

2. Performance Metrics and Targets

This is the core of the SLA. Common metrics include:

• Availability / Uptime: Percentage of time a system is operational. 99.9% = ~8.7 hours downtime/year. 99.99% = ~52 minutes.
• Response Time: How quickly the provider acknowledges an issue (not resolves it).
• Resolution Time: How quickly an issue is fully resolved.
• Throughput: Transactions per second, requests handled, etc.
• Error Rate: Acceptable percentage of failed requests.

3. Severity Levels

Define a tiered severity system:

Severity	Definition	Response Time	Resolution Time
P1 (Critical)	System down, no workaround	15 minutes	4 hours
P2 (High)	Major feature impaired	1 hour	8 hours
P3 (Medium)	Minor feature impaired	4 hours	24 hours
P4 (Low)	Cosmetic, enhancement	Next sprint	—

4. Measurement and Reporting

How will compliance be measured? Who measures it? When are reports delivered? Monthly SLA reports are standard. Define exclusions clearly — scheduled maintenance windows, force majeure events, and customer-caused outages typically don't count against uptime.

5. Remedies and Credits

What happens when the SLA is missed? Service credits are standard — usually a percentage of the monthly fee per incident:

• Miss by 0.1%–1%: 5% monthly credit
• Miss by 1%–5%: 15% monthly credit
• Miss by >5%: 25–50% monthly credit or termination right

Credits are not the same as refunds. Define whether credits are automatic or require a claim within a specified window.

6. Escalation Procedures

Define exactly how issues escalate: who gets notified at each severity level, how quickly escalations happen, and who the senior contacts are on both sides.

7. Exclusions

List what's explicitly NOT covered: client-caused issues, force majeure, third-party dependencies (internet providers, DNS, etc.), scheduled maintenance windows, and services outside the agreed scope.

Common SLA mistakes to avoid

Setting targets you can't measure. If you commit to 99.9% uptime but don't have monitoring infrastructure that tracks it, you've created a liability you can't defend.

No escalation path. "Contact support" is not an escalation path. Name specific individuals or roles with contact information.

Credits without a claim process. Automatic credits are customer-friendly but operationally complex. Most SLAs require customers to submit claims within 30 days.

Ignoring the exclusions. The exclusions list is as important as the commitments. An SLA without clear exclusions will generate disputes every time a third-party outage causes downtime.

Generate a free SLA template →