Skip to content
← All articlesCloud & DevOps

SLOs and Error Budgets: A Guide for Product Teams

Learn how SLIs, SLOs and error budgets turn vague uptime promises into clear targets, and how to agree them with an outsourced development team.

“The app must be available all the time” is one of the most common requirements in a software contract, and one of the least useful. It cannot be measured, it cannot be costed, and it gives the team no way to decide between shipping a feature and fixing a reliability problem.

Service level objectives (SLOs) and error budgets replace that sentence with something concrete. They give founders, product managers and engineers a shared language for how reliable a service needs to be, and what to do when it falls short.

The three terms, in plain language

  • SLI (service level indicator): a measurement of how the service behaves, such as the share of requests that succeed or the share that complete within a set time.
  • SLO (service level objective): the target for that measurement over a period, for example “99.9% of checkout requests succeed over 30 days”.
  • SLA (service level agreement): a contractual commitment to a customer, usually with remedies if it is missed. An SLA should be looser than the internal SLO, so you have room to react before a penalty applies.

The error budget is simply what is left after the SLO: if the target is 99.9% over 30 days, the remaining 0.1% of that period is the amount of failure you can tolerate. A 99.9% target over 30 days allows roughly 43 minutes of full unavailability.

Why error budgets change the conversation

Without a budget, reliability and speed of delivery are argued about on opinion. With one, the rule is simple:

  • Budget remaining: ship features at normal pace.
  • Budget nearly spent: slow down risky releases and prioritise stability work.
  • Budget exhausted: freeze non-essential releases until reliability recovers.

This removes the blame from the discussion. Nobody is “too cautious” or “too reckless”; the data decides.

Choosing good SLIs

Measure what users actually experience, not what servers report. A server can be healthy while the login page fails.

User journey Useful SLI
Browsing or API reads Percentage of requests returning a non-error response
Checkout or payment Percentage of attempts that complete successfully
Search or dashboards Percentage of requests faster than an agreed threshold
Background processing Percentage of jobs finished within the agreed window
Mobile app launch Percentage of sessions that start without a crash

Pick two to four SLIs for the journeys that matter most to revenue or retention. More than that and nobody will look at them.

Setting realistic targets

Higher targets cost disproportionately more. Moving from 99% to 99.9% typically needs better monitoring, automated rollbacks and redundancy; going further often needs multi-region design and a larger on-call commitment.

A practical approach:

  1. Start from current performance. Measure for a few weeks before promising anything.
  2. Set the first SLO slightly below what you already achieve.
  3. Ask what an hour of downtime costs the business, and what the user will tolerate.
  4. Review quarterly and tighten only when users or revenue justify it.

Not every feature needs the same target. A payments flow and an internal reporting page deserve different objectives.

Putting SLOs to work with an outsourced team

SLOs are especially valuable when development and operations are split across companies or time zones, because they replace subjective status reports with numbers both sides trust.

  • Write them into the support agreement. Link SLOs to response and resolution times in your application support SLA.
  • Alert on budget burn, not every blip. Page someone when the budget is being consumed quickly, and send slower burns to a ticket queue.
  • Review monthly. Look at budget spent, the incidents behind it and the follow-up actions agreed.
  • Feed incidents back. Each breach should end in a short review and a concrete fix, as described in our guide to incident management for live applications.

A simple starting checklist

  • List the three to five user journeys that matter most.
  • Define one or two SLIs for each, using data you can already collect.
  • Agree an SLO and a measurement window (30 days is common).
  • Build a dashboard showing remaining error budget.
  • Define what happens at each stage of budget consumption.
  • Assign an owner for the monthly review.

Common mistakes

  • Setting 100% as the target. It is unachievable and leaves no budget for change.
  • Measuring only server uptime and ignoring user-facing errors and latency.
  • Copying another company’s targets instead of using your own data.
  • Creating SLOs that nobody reviews, so they become decoration.

Next steps

If you want reliability targets defined, monitored and enforced across your product, our cloud and DevOps services cover monitoring, alerting and release automation, and our support and maintenance teams run live applications against agreed objectives. To discuss your own setup, contact us.

Let’s build what’s next.

Tell us about your platform, your product idea or the systems you want to modernise. Someone from our Chennai engineering team will get back to you personally.