A service level indicator, or SLI, is a measured value of service performance as users experience it. Common examples are the share of requests that succeed, the time taken to respond and the proportion of data that is fresh. An SLI is the measurement itself, which is then compared with a target, called a service level objective, to judge whether the service is meeting expectations.
Choosing good SLIs keeps attention on what users actually notice. A server that reports healthy while users receive errors is not meeting its purpose, and measuring from the user's side, or close to it, avoids that blind spot.
Details matter. Averages can hide a slow tail, so percentiles such as the 95th or 99th are often more informative for response times. Where and how the measurement is taken changes the number, and a metric that is easy to collect is not necessarily the one that reflects user experience.
To define one, pick a small number of indicators tied to key user journeys, write down exactly how each is measured, set targets together with the people who depend on the service and review them as usage and expectations change. Do not confuse an indicator with a contractual service level agreement.