Free tool
Uptime SLA calculator
Convert a target uptime percentage into the actual downtime budget it allows — per hour, day, month, quarter, and year. Use the slider or type any value to explore the math behind "three nines" vs "four nines."
Operational tier
Three nines
Realistic ceiling for most SaaS on a single cloud region with load balancing and on-call rotation.
Allowed downtime
| Per hour | 4 seconds |
| Per day | 1m 26s |
| Per week | 10m 5s |
| Per month (30 days) | 43m 12s |
| Per quarter (91 days) | 2h 11m 2s |
| Per year (365 days) | 8h 45m 36s |
How to read the result
The downtime numbers above are an absolute budget — they include every minute the service is unavailable, regardless of cause: deploys, database failovers, third-party outages, certificate renewals, anything. Most published SLAs exclude planned maintenance and force-majeure events, so your publishable SLA tends to be one tier easier than your raw measured availability.
The rule of thumb: each extra nine costs you roughly 10× the downtime budget — and 10× the operational investment to back it up. We wrote a deeper guide on what each tier actually costs in practice.
Want to measure your real uptime?
The calculator tells you what your target permits. To find out whether you're actually hitting it, you need independent multi-region monitoring measuring availability separately from your own infrastructure. That's exactly what Uptimera does — 5 monitors free, forever.