Uptime SLA Calculator: Downtime Per Day, Month, and Year

Enter an uptime percentage and see how much downtime that SLA actually allows per day, week, month, and year. Or flip it: enter the downtime you had and find out what percentage you hit. Free, no signup, runs in your browser.

Reverse: downtime to uptime %

Had an incident and need to report the resulting uptime? Enter total downtime and the period it occurred in.

Downtime allowed at each level of nines

The standard tiers, precomputed. A 30-day month is used because that's what most SLAs and monitoring tools measure against, not the calendar month.

What each uptime tier means, in plain terms

Uptime is usually quoted in "nines," and each nine you add cuts the allowed downtime by a factor of ten. Here's what the common tiers work out to against a 30-day month:

• 99% (two nines) allows 7 hours 12 minutes of downtime a month. That sounds generous until you realize it's most of a working day offline every month.
• 99.9% (three nines) allows 43 minutes 12 seconds a month, or about 8 hours 46 minutes a year. This is the most common target for small SaaS products and the default most people mean by "high uptime."
• 99.95% allows 21 minutes 36 seconds a month. A frequent middle ground for paid business tiers.
• 99.99% (four nines) allows 4 minutes 19 seconds a month, or 52 minutes 34 seconds a year. Reaching this reliably means redundancy, fast failover, and detection measured in seconds rather than minutes.
• 99.999% (five nines) allows 26 seconds a month. This is a hardware-redundancy, multi-region, no-single-point-of-failure commitment, and it's expensive to hold honestly.

SLA, SLO, and error budget: the difference

These three terms get used interchangeably and they shouldn't be. An SLA (service level agreement) is the contractual promise you make to customers, usually with financial penalties attached if you miss it. An SLO (service level objective) is the internal target you actually aim for, and it's almost always stricter than the SLA so you have a safety margin before you owe anyone money. The error budget is the gap the SLO leaves you: if your SLO is 99.9%, your error budget is the 0.1% of downtime you're allowed to spend before you're at risk. When the budget is healthy you can ship risky changes; when it's nearly gone, reliability work takes priority. The error budget calculator works this out from the same numbers.

Why a 30-day month, not a calendar month

This calculator uses a fixed 30-day month (2,592,000 seconds) rather than the actual calendar length. That's deliberate, and it's how most monitoring tools and vendor SLAs compute it, because a fixed window makes months comparable and keeps the math consistent. A February incident and an August incident of the same length should register the same, and a 30-day basis is the common convention that makes that true. If your vendor measures per calendar month or per billing cycle instead, the allowed-downtime figures shift slightly, so it's worth checking which basis a contract actually uses.

What is a good uptime SLA target?

There's no universal answer; the right target depends on what breaks when you're down and what redundancy costs to add. For most small SaaS products and internal tools, 99.9% is the sensible target: it's achievable on a well-run single-region setup with good monitoring, and it maps to a downtime budget you can actually defend. Pushing to 99.99% roughly multiplies your infrastructure and operational cost because it forces redundancy and near-instant failover, so it's worth committing to only when the revenue or contractual stakes justify it. The honest move is to set an SLO you can consistently hit, measure against it, and raise it once you're comfortably beating it, rather than promising nines you can't hold.

Notes worth knowing

An SLA number on its own says nothing about incident shape. 43 minutes per month at 99.9% could be one painful outage or thirty short blips, and your users experience those very differently. Detection time eats into the budget too; if your monitoring checks every 5 minutes, you can lose up to 5 minutes of budget before anyone even knows.

Watch the measurement window in vendor SLAs. Some compute uptime per calendar month, some per billing cycle, and a few measure per year, which lets a vendor burn the entire annual allowance in one incident while technically staying compliant.

For a fuller walkthrough of turning a target into a budget you can track week to week, see the guide on how much downtime 99.9% uptime allows.