What is the difference between an SLA and an SLO?

An SLA is the external, usually contractual promise made to customers, with consequences such as service credits if it is missed. An SLO is the internal target a team sets for itself, typically stricter than the SLA so there is a safety buffer before the contract is breached.

How much downtime does a 99.9% SLA allow?

About 43 minutes per month, or roughly 8.8 hours per year. Tighter targets allow far less: 99.99% permits around 4.3 minutes a month, and 99.999% ("five nines") only about 26 seconds.

What happens if a provider breaches its SLA?

The agreement specifies the remedy, most commonly a service credit applied to a future invoice and scaled to how far below target the service fell. SLAs rarely refund cash or cover the customer's downstream losses; the credit is usually the customer's sole contractual recourse, and it normally has to be claimed within a set window.

Does scheduled maintenance count against an SLA?

Usually not. Most SLAs exclude pre-announced maintenance windows from the availability calculation, along with outages caused by the customer or by third parties outside the provider's control. The exact exclusions are defined in the agreement, so they are worth reading closely.

What Is an SLA (Service Level Agreement)?

Reviewed by Ionut Caval · Updated June 2026

A Service Level Agreement (SLA) is a formal, usually contractual commitment between a provider and its customers that defines the expected level of service, such as 99.9% uptime, and the remedies (like service credits) that apply if that level is not met.

An SLA turns a reliability promise into something measurable and accountable. Instead of a vague claim that a service is "highly available", it commits the provider to a specific target, states exactly how that target is measured, and spells out what the customer is owed when it is missed. SLAs are most common in B2B software, hosting, and cloud infrastructure, where downtime has a direct cost for the customer's own business.

What an SLA defines

A complete SLA goes well beyond a single percentage. It typically specifies:

The service-level target, for example 99.9% uptime over a calendar month.
How availability is measured, and from where (often externally, the way a real visitor connects).
What counts as downtime and what is excluded, such as scheduled maintenance windows.
The remedy if the target is missed, usually a service credit scaled to how badly it was breached.
How the customer claims that remedy, and within what time limit.

A worked example

The headline number is easy to underestimate. A 99.9% monthly uptime SLA still permits about 43 minutes of downtime every month; 99.99% tightens that to roughly 4.3 minutes, and 99.95% sits in between at about 22 minutes. A single longer outage can therefore exhaust an entire month's budget at once. For a fuller breakdown of each "nine", see what 99.9% uptime actually means, or convert any percentage into allowed minutes with the uptime and SLA calculator.

On the remedy side, a typical tiered SLA might credit 10% of the monthly fee if uptime falls below 99.9%, 25% below 99.0%, and 50% below 95.0%. So on a $1,000/month plan that dropped to 98.7% availability, the customer would be owed a $250 credit.

The SLA is the customer-facing layer of reliability. Internally, teams manage toward an SLO (the target they aim for) measured by an SLI (the actual metric), and they usually set the SLO tighter than the SLA so a bad week does not immediately breach the contract. Continuous external SLA monitoring is how you know in real time whether you are still inside the agreed budget.

Frequently asked questions

What is the difference between an SLA and an SLO?

An SLA is the external, usually contractual promise made to customers, with consequences such as service credits if it is missed. An SLO is the internal target a team sets for itself, typically stricter than the SLA so there is a safety buffer before the contract is breached.
How much downtime does a 99.9% SLA allow?

About 43 minutes per month, or roughly 8.8 hours per year. Tighter targets allow far less: 99.99% permits around 4.3 minutes a month, and 99.999% ("five nines") only about 26 seconds.
What happens if a provider breaches its SLA?

The agreement specifies the remedy, most commonly a service credit applied to a future invoice and scaled to how far below target the service fell. SLAs rarely refund cash or cover the customer's downstream losses; the credit is usually the customer's sole contractual recourse, and it normally has to be claimed within a set window.
Does scheduled maintenance count against an SLA?

Usually not. Most SLAs exclude pre-announced maintenance windows from the availability calculation, along with outages caused by the customer or by third parties outside the provider's control. The exact exclusions are defined in the agreement, so they are worth reading closely.

Put these metrics to work. Monitor your site free.

2-minute setup · Cancel any time
Create Free Account

Your Dashboard

No credit card needed