Use this uptime calculator to translate 99%, 99.9%, 99.95%, 99.99%, five nines, or any custom SLA target into allowed downtime per day, week, month, quarter, and year.
Set an uptime target and compare the downtime budget across useful windows. The calculator uses 24-hour days, 7-day weeks, 30-day months, 90-day quarters, and 365-day years.
Uptime is the percentage of a measurement window where a service is available. If a service is available for 29 days, 23 hours, and 16 minutes in a 30-day month, its uptime is the available time divided by the full month.
The inverse is often more useful for planning: decide the uptime target first, then calculate the downtime budget it permits. That is the number on-call teams, customer success teams, and status page owners can reason about during incident planning.
| Target | Name | Downtime / year | Downtime / month | Typical fit |
|---|---|---|---|---|
| 99% | Two nines | 3d 15h 36m | 7h 12m | Internal tools |
| 99.9% | Three nines | 8h 45m 36s | 43m 12s | Most SaaS products |
| 99.95% | Business critical | 4h 22m 48s | 21m 36s | Revenue-sensitive apps |
| 99.99% | Four nines | 52m 34s | 4m 19s | Platform services |
| 99.999% | Five nines | 5m 15s | 26s | Critical infrastructure |
SLA language matters. Some teams count any failed check as downtime. Others count only user-impacting outages, slow responses beyond a threshold, or global failures confirmed from more than one monitoring region. For transparent reporting, define these rules before an incident happens.
Scheduled maintenance is another important distinction. If planned maintenance is excluded from your SLA, your monitoring and reports should separate maintenance windows from unplanned downtime.
A yearly average can hide a bad month. A service can look healthy annually while breaching a monthly customer SLA. Pick the same measurement window your contract, status page, or internal objective uses, then calculate uptime against that window.
Server logs can show that your process was running, but customers can still be blocked by DNS errors, certificate failures, regional routing problems, CDN issues, or slow responses. External uptime monitoring gives you the user-facing availability signal that SLA math depends on.
Uptime percentage equals available time divided by total time, multiplied by 100. Equivalently, subtract downtime from the measurement window before dividing by the full window.
Using a 365-day year, 99.9% uptime allows about 8 hours, 45 minutes, and 36 seconds of downtime per year, or about 43 minutes and 12 seconds in a 30-day month.
That depends on the SLA. Many SLAs exclude communicated maintenance windows, but public uptime reports should clearly explain whether planned maintenance is included.
MarionetteOps gives you free uptime monitors, server agents, public status pages, and alert routing so your uptime target is backed by real checks.