SLO (Service Level Objective)

An SLO is built from an SLI (Service Level Indicator) — the actual metric being measured, like request latency or success rate — plus a target and a time window. “99.9% of requests succeed in under 300ms, measured over a rolling 30 days” is a complete SLO: it names the metric, the threshold, and the period.

The number that matters day to day is what’s left over: the error budget. If the target is 99.9%, the budget is the 0.1% of requests allowed to fail or be slow. As long as a team is within budget, they can ship features and take on risk. Once the budget is nearly spent, the team shifts focus to stability — not because a manager said so, but because the data says so.

SLOs are different from SLAs (Service Level Agreements) — an SLA is a contractual promise to a customer, often with financial penalties attached, while an SLO is typically an internal target set stricter than the SLA to leave a safety margin. A team might target a 99.95% internal SLO to comfortably meet a 99.9% SLA promised to customers.