A service level agreement template is a ready-made contract that sets out the service a provider must deliver and how it will be measured. It fixes uptime, response and resolution targets, and the service credits owed when the provider misses them.
An SLA template is the starting draft both sides use to agree, in measurable terms, what good service looks like. Instead of promising to be fast or reliable, it states a number, explains how that number is measured, and says what the customer gets back if it is missed.
Buyers use it when signing up a software, cloud, IT support or outsourced service provider, usually as a schedule to the main services agreement. Providers use the same structure to publish standard terms, and internal teams such as IT use it to set service levels for the rest of the business.
Because the SLA is where service failures turn into money, it is the part of a contract most often negotiated after price.
Who the parties are, when the SLA takes effect, how long it runs and what it is for.
The tasks, systems and support covered, and just as clearly, what is excluded.
Measurable targets such as uptime percentage, error rate and reliability.
How fast the provider must acknowledge and fix issues, split by priority level.
Service credits or refunds owed when a target is missed.
Sign-off by an authorised person for both the client and the provider.
Ready to use in Word, Google Docs and PDF. Fill it in, save it, reuse it.
Agree the services in scope, the metrics and the target for each one.
The provider tracks uptime and ticket times with its monitoring and support tools.
A monthly report shows each metric against target, with incidents listed.
Missed targets trigger service credits, which the customer usually has to claim within a set period.
Both sides revisit targets each quarter or at renewal as the service changes.
The simplest usable SLA has four lines: the service, the uptime target, the support response time and the credit if either is missed. Add scope, priorities and reporting as the contract grows.
Clauses 1 to 5 and 9 follow the order most SLAs use: overview, scope, metrics, response times, then remedies. The clauses in between deal with how service is measured and what the customer must do. Delete any clause that does not apply rather than leaving it blank.
Parties, effective date, term, purpose, and the main agreement this SLA sits under.
Services, systems and locations covered, plus explicit exclusions.
When support is available, by which channel, and in which time zone.
Each metric, its target, and exactly how and over what period it is measured.
Targets by priority level, from critical outages to minor requests.
What the monthly report contains, when it is due, and the quarterly review meeting.
Access, contacts and information the customer must provide for targets to apply.
Planned maintenance windows, customer-caused issues and events outside either party's control.
Service credit levels, how to claim them, and the cap on total credits.
The customer's right to terminate if targets are missed repeatedly.
How metrics or targets can be changed, and that changes need written agreement.
Name, title, signature and date for an authorised person at each party.
1. Agreement Overview
This Service Level Agreement ("SLA") is made on [Effective Date] between [Client Name] ("Client") and [Provider Name] ("Provider"). It forms part of the [Master Services Agreement] dated [Date] and applies to the services listed in Schedule A for [12 months], renewing with that agreement.
9. Penalties and Remedies
If Monthly Uptime falls below the target in Clause 4, Provider will credit Client the percentage of the monthly fee shown in Schedule C. Client must request the credit within [30] days of the monthly report. Total credits in any month will not exceed [50]% of that month's fee, and credits are applied to the next invoice.
12. Signatures
Signed for [Client Name]: [Name], [Title], [Signature], [Date]
Signed for [Provider Name]: [Name], [Title], [Signature], [Date]Uptime is the headline metric for software and cloud services, while support SLAs lean on response and resolution times. For each metric, write down the data source, the measurement period and what is excluded. Targets below are illustrative, not market standards.
| Metric | Definition | Example target | Measured by |
|---|---|---|---|
| Uptime | Minutes the service is available ÷ total minutes in the month | 99.9% monthly | Provider's external monitoring |
| Error rate | Failed requests ÷ total requests | Below 0.5% | Application logs |
| First response time | Time from ticket logged to first human reply | By priority, see next part | Support desk timestamps |
| Resolution time | Time from ticket logged to fix confirmed | By priority, see next part | Support desk timestamps |
| Data backup | Successful daily backups ÷ scheduled backups | 100% | Backup job reports |
Based on a 30-day month of 43,200 minutes.
| Uptime target | Allowed downtime per month | Allowed downtime per year |
|---|---|---|
| 99.5% | 216 minutes (3.6 hours) | 43.8 hours |
| 99.9% | 43.2 minutes | 8.8 hours |
| 99.95% | 21.6 minutes | 4.4 hours |
| 99.99% | 4.3 minutes | 52.6 minutes |
| Item | Formula |
|---|---|
| Monthly uptime (total minutes in B2, downtime in C2) | =(B2-C2)/B2 |
| Allowed downtime for a target in D2 | =B2*(1-D2) |
Response time is how quickly the provider acknowledges the issue; resolution time is how quickly it is fixed or worked around. Agree who sets the priority, and allow the customer to raise it if the impact grows.
| Priority | Definition | Response | Resolution | Updates |
|---|---|---|---|---|
| P1 Critical | Service down or data at risk for all users | 30 minutes, 24/7 | 4 hours | Every hour |
| P2 High | Major function impaired, no workaround | 1 hour, 24/7 | 8 hours | Every 4 hours |
| P3 Medium | Partial loss, workaround available | 4 business hours | 3 business days | Daily |
| P4 Low | Question, request or cosmetic issue | 1 business day | 10 business days | Weekly |
Illustrative targets. Tighten P1 and P2 for business-critical services.
Credits are the usual remedy because they are simple to calculate and avoid arguments about actual loss. Most SLAs say credits are the customer's main remedy for missed service levels, so check what other rights you keep.
| Monthly uptime | Credit (% of monthly fee) |
|---|---|
| 99.9% or above | 0% |
| 99.0% to below 99.9% | 10% |
| 95.0% to below 99.0% | 25% |
| Below 95.0% | 50% |
Illustrative credit tiers; agree your own with the provider.
| Item | Formula |
|---|---|
| Credit due (uptime in B2, monthly fee in C2) | =IF(B2>=0.999, |
Illustrative figures; Kestrel Data is fictional.
Most buyers sign a service-based SLA, because the provider offers the same terms to every customer. Large customers negotiate customer-based terms, and internal IT teams often use a multi-level structure.
One SLA for one customer covering every service they buy, such as a bank's full IT outsourcing contract.
One SLA for one service, offered on the same terms to all customers, such as a cloud provider's uptime terms.
Layers of terms: corporate-wide levels, then customer-specific and service-specific levels on top.
SERVICE LEVEL AGREEMENT Between [Client Name] and [Provider Name], effective [Date], under the agreement dated [Date]. Service: [Describe the service in one sentence]. Availability: [99.9]% each calendar month, excluding planned maintenance notified [48] hours ahead. Support: [Email / phone], [09:00-18:00] [time zone], Monday to Friday. Response: Critical issues within [1] hour; all others within [1] business day. Service credit: [10]% of the monthly fee for any month below the availability target. Review: Every [3] months. Signed: [Client name, title, date] [Provider name, title, date]
Spendflo's Contracts Agent stores every SLA with its contract and tracks each renewal date.
See the Contracts AgentThe share of the measurement period the service is available, after agreed exclusions.
Time the service is unavailable to users, as detected by the agreed monitoring.
Time from the customer logging an issue to the provider acknowledging it.
Time from logging to a fix or acceptable workaround being confirmed.
A reduction in fees owed to the customer when a target is missed.
Pre-announced time for planned work that does not count as downtime.
| Document | Answers | Example content |
|---|---|---|
| Master services agreement | On what legal terms do we work together? | Liability, confidentiality, payment, termination |
| Statement of work | What exactly will be delivered? | Deliverables, milestones, fees |
| Service level agreement | How well must the service perform? | Uptime, response times, service credits |
For software buyers, see how SaaS service level agreements differ from other services.
Name the data source, period and exclusions next to every target.
A P1 should mean users cannot work, not that someone escalated loudly.
Credits are hard to claim without the provider's own figures.
Repeated misses should let you leave without a termination fee.
Usage and criticality change, so the SLA should change with them.
"Best efforts" and "high availability" cannot be measured or enforced.
Cap planned downtime and require advance notice, or uptime targets become meaningless.
Short claim windows mean most credits are never paid.
A fast acknowledgement with a slow fix still leaves users stuck.
List what is covered and what is not, including systems, locations and hours.
Choose three to five metrics and agree how each is measured.
Use the priority and credit tables above as your first draft.
Fix the report date and review cadence, then sign alongside the main contract.
Harbour Facilities signs a helpdesk SLA with Northwind IT at 6,500.00 a month. P1 tickets need a response in 30 minutes and a fix in 4 hours. In June, two P1 fixes take 7 hours, triggering the agreed 10% credit of 650.00 on the July invoice. All figures and names are illustrative.
Every part on this page, in Word, Google Docs and PDF, with the examples filled in.
Lead with monthly uptime, maintenance windows and data backup, and check the provider's standard credit tiers before you negotiate.
Lead with response and resolution times by priority, staffing hours and reporting. Read more on managing a vendor SLA.
IT or shared services set targets for other departments, with escalation paths instead of service credits.
Best for negotiation and redlines.
Best for joint drafting.
Best for the signed, final copy.
Spendflo has handled 15,000+ agreements, at 30% average savings on software spend.
See your savingsA good SLA turns service quality into numbers, measures them monthly and pays out when they slip. The hard part comes after signing: reading the reports, claiming credits and using the record at renewal.
Quick answers to what people ask most about the service level agreement template.
Define the services in scope, set measurable targets such as uptime and response times by priority, and attach service credits for missed targets. Add reporting, exclusions and signatures. Download the template above to start from all twelve clauses.
A common example is a cloud provider promising 99.9% monthly uptime, with a 10% fee credit if it falls short. Another is a helpdesk promising a 30-minute response to critical tickets. The download includes both kinds of targets with sample wording.
The three types are customer-based, service-based and multi-level SLAs. Customer-based covers every service for one customer, service-based covers one service for all customers, and multi-level layers both. Download the template to adapt it to any of the three.
Keep it to one page: the parties, the service, one availability target, support hours, a response time, a service credit and signatures. Download the template and use the simple one-page version in Part 5, then have counsel review it.
You can download this free SLA template in Word, Google Docs or PDF using the buttons on this page. It includes the full agreement, metric and priority tables, service credit tiers and a one-page version.
Contracts
Purchase orders
Vendor management
Sourcing and RFx
Budgets and business cases
Procurement
Accounts payable
Purchasing
Software buying
Supply chain
Spendflo routes new service purchases through intake and approvals, and its Contracts Agent stores each agreement and tracks renewal dates.
Enter your work email and we'll unlock every format.
Didn't start, or need another format? Pick one below.
Google Docs: upload the file to Google Drive, then open it with Google Docs.
