SLA uptime calculator

Downtime allowed by common SLA targets

A month is an average month (30.44 days), a quarter is 91.31 days and a year is 365.25 days, so leap years even out.

How the calculation works

An uptime percentage is the share of a period a service must be available. The downtime it allows is the rest: allowed downtime = (100 − uptime) ÷ 100 × the length of the period. At 99.9% that's 0.1% of the time, which is 1 minute 26 seconds a day or about 43 minutes 50 seconds in an average month. Each extra nine divides the allowance by ten: 99.99% leaves about 4 minutes 23 seconds a month, and 99.999% about 26 seconds.

Working backwards, uptime = 100 − (downtime ÷ length of the period × 100). A 45-minute outage in a 30.44-day month leaves about 99.897% uptime, just under a 99.9% target.

What an SLA percentage doesn't tell you

The same percentage can mean very different things, so read the definitions in any agreement you sign or offer:

• The period. 99.9% a month allows one 43-minute outage each month. 99.9% a year allows nearly nine hours at once.
• What counts as down. A full outage only, or also errors and very slow responses? Some agreements count partial outages at a fraction.
• Planned maintenance. Many agreements leave scheduled maintenance out, provided it's announced in advance.
• Who measures. Your own monitoring, the customer's, or a third party's.

Measuring uptime against a target

You can only report on downtime your monitoring can see. A check every 5 minutes can miss a two-minute outage entirely, and can overstate a short one by up to five minutes. For 99.9% and above, check every minute or faster: Uptime Dog checks as often as every 30 seconds on Essentials and Pro, and every 20 seconds on Business.

Uptime Dog reports uptime, downtime, incident count, mean time to recover and the longest incident for any period, with CSV export, and shows 30, 60 or 90 days of uptime bars on your status page. Maintenance windows are recorded separately, so planned work doesn't count against you. Getting outages fixed quickly matters as much as preventing them, which is where on-call schedules and escalation come in.