What is an SLA?
An SLA is a written promise from a provider about the level of service, such as 99.9% monthly uptime, and what you get, often a service credit, when it is missed.
In more detail
A useful SLA names four things: the measure (uptime, response time), the period (usually a calendar month), the exclusions (planned maintenance, problems on your side) and the remedy (usually a credit of part of the monthly fee, which you have to claim).
The provider measures its own SLA. Independent monitoring gives you your own record of when the service was down, which is what you need to make a claim.
Worked example
A host promises 99.9% monthly uptime and a 10% credit below it. Your monitoring recorded two hours of outages in a 30-day month: 1 − 120 ÷ 43,200 = 99.722%. That is below 99.9%, so you have a claim, unless the outages fall under an exclusion.
See also
Related terms: SLO (service level objective) and Uptime. Free tool: Uptime calculator. How OwlHound helps: Website uptime monitoring. All terms: monitoring glossary.
Frequently asked questions
What is the difference between an SLA and an SLO?
An SLA is a promise to a customer with a remedy attached. An SLO is an internal target, usually stricter, so a team notices problems before the SLA is at risk.
How do I check whether my host met its SLA?
Keep your own record. An outside monitor logs every confirmed outage with its start and end, and a monthly uptime figure to compare with the promise.
What does a 99.9% SLA allow?
About 43 minutes of downtime in a 30-day month.
Start watching your site in minutes
Free for up to 10 sites, no card, nothing to install. Read-only access you can revoke at any time — never passwords, never admin access.