What Is an SLO (Service Level Objective)?

Reviewed by Ionut Caval · Updated June 2026

A Service Level Objective (SLO) is a precise, internal reliability target that an engineering team commits to managing toward, for example "99.95% of requests succeed over a rolling 30-day window" or "99% of responses return in under 300 ms." It is the goal you steer by, sitting between the metric you measure and the promise you make to customers. An SLO has three parts: an SLI (the metric being measured), a target value (the percentage or threshold), and a time window (rolling 28 or 30 days is common). Without all three it is not actionable, since 99.9% means nothing until you say 99.9% of what, measured over how long.

SLO vs SLA: why the SLO is tighter

The SLO is internal; the SLA is the external, often contractual promise to customers, with credits or penalties when it is missed. Teams deliberately set the SLO stricter than the SLA so there is a buffer. If your SLA guarantees 99.9% uptime (about 43 minutes of downtime per month), you might run an internal SLO of 99.95% (about 22 minutes). The 21-minute gap is your safety margin: a bad week can burn the SLO without breaching the customer SLA. For a full breakdown of what each "nine" of uptime allows, see what 99.9% uptime actually means.

Error budgets

The flip side of an SLO is the error budget: the small amount of failure the target permits. A 99.95% availability SLO leaves a 0.05% budget, roughly 22 minutes of allowed downtime in a 30-day month. Error budgets turn reliability into a spendable resource:

• When the budget is healthy, teams can ship faster and take more risk, since there is room to absorb a small failure.
• When the budget is nearly exhausted, the sensible response is to slow releases and prioritise reliability work until it recovers.
• A consistently overspent budget signals the SLO is too strict for current capacity, or that the system needs real investment.

Picking realistic SLOs and measuring the underlying SLIs from outside your infrastructure, the way a real visitor connects, is what keeps the target honest. Uptime and SLA monitoring records the availability data you need to track an SLO and prove an SLA.

See also: Uptime & SLA monitoring