An SLI is a measured indicator of service behavior, such as the fraction of requests served under 300ms. An SLO is the internal target for that indicator, for example 99.9% of requests under 300ms over 28 days. An SLA is the external contract with customers, usually carrying financial penalties, and it should be looser than your SLO so you have headroom before you owe anyone money.
Why interviewers ask this
The interviewer wants to know whether you can turn vague reliability talk into numbers a team can act on. Mixing up the three terms signals you have never owned a service in production. They are also listening for whether you tie the SLO back to user experience rather than infrastructure health, and whether you understand that an SLA is a commercial artifact, not an engineering target.
How to structure your answer
- Define all three in one clean sentence each, in order of scope.
- Anchor the SLI to something the user actually feels, not CPU or disk.
- State the measurement window and where the number is measured from.
- Close on why the SLA must be looser than the SLO.
Example answer
So an SLI is the measurement, an SLO is the target, and an SLA is the promise you sell. On my last team the SLI for our checkout API was the proportion of requests that returned a non 5xx status in under 400ms, measured at the load balancer rather than inside the app, because that is closer to what the user sees. The SLO was 99.9% of those requests over a rolling 28 day window. Our SLA with enterprise customers was 99.5% monthly with service credits attached, deliberately looser, so we had a whole band of degradation to absorb before it became a commercial problem. The bit people miss is that the SLI has to be a user journey. We had a beautiful CPU dashboard and it never once predicted an outage. Once we switched to request based indicators, we could actually argue about priorities with product using the same numbers.
Walking into this interview soon? GhostPilot listens to your live call, spots the question the moment it is asked, and puts a structured answer on your screen in real time. Try it on your next mock, or grab a $29 Session Pass, no subscription, for the real thing.
See how it worksFollow-up questions to expect
- Where exactly would you measure that SLI, and what does that choice miss?
- How would you set the initial target if you have no historical data?
- What would you do if the business wants an SLA tighter than your SLO?
Related site reliability engineer questions
Your interviewer will ask their own version of this. Paste your actual job description into the free Question Predictor and get the 20 questions that role is most likely to ask, with what each one is really probing.
Predict my questions