Free tool · Calculator

SLA & Uptime Calculator

Turn an availability target into the downtime it really allows, work backwards from an outage, and see what your dependencies do to end-to-end availability.

  • Downtime per target
  • Delivered availability
  • End-to-end chains

A 99.9% availability target on a 24×7 service allows 8h 45m 36s of downtime a year, or 43m 48s a month. Add one more nine, 99.99%, and the monthly allowance drops to 4m 23s. Measured only over a Sunday–Thursday week of 9-hour days, 99.9% allows 11m 44s a month. Figures use a 365-day year and a month of one-twelfth of a year.

1. How much downtime does my target allow?

Enter an availability target and the hours the service must be up.

2. What availability did we actually deliver?

Enter the downtime in a period to get the availability you achieved.

3. End-to-end availability across dependencies

A service is only as available as the chain it depends on. Add each component, and tick redundant if a duplicate can take over.

Each link is a promise someone else makes you: an OLA from an internal team or an underpinning contract from a supplier. The SLA you sign can never be stronger than the chain behind it.

Availability reference table

Allowed downtime for common targets on a 24×7 service.

AvailabilityPer yearPer quarterPer monthPer weekPer day
99%Two nines3d 15h 36m21h 54m7h 18m1h 40m 48s14m 24s
99.5%1d 19h 48m10h 57m3h 39m50m 24s7m 12s
99.9%Three nines8h 45m 36s2h 11m 24s43m 48s10m 5s1m 26s
99.95%4h 22m 48s1h 5m 42s21m 54s5m 2s43s
99.99%Four nines52m 34s13m 8s4m 23s1m9s
99.999%Five nines5m 15s1m 19s26s6s1s

Writing an availability SLA that holds up

  • Define what is measured. Name the service and the point of measurement, such as the user login page, not a single server.
  • State the service hours and window. 24×7 or business hours, measured monthly or quarterly. The same percentage means very different downtime in each.
  • Agree the exclusions up front. Planned maintenance with notice, customer-caused outages and force majeure, so a breach is never a debate.
  • Say where the numbers come from. Name the monitoring source and how partial outages or degraded performance count.
  • Back it with OLAs and supplier contracts. If three dependencies each promise 99.9%, you cannot credibly promise 99.9% end to end. Use calculator 3.
  • Make breaches mean something. Agree the reporting cadence, service credits and what triggers a service improvement plan. The downtime cost calculator puts a price on each target.

Method & ownership

Allowed downtime is the period length × (1 − availability), with a 365-day year. Business-hours coverage scales the period by hours per day × days per week out of 24 × 7. End-to-end availability multiplies the components in series; a redundant pair counts as 1 − (1 − A)².

The formulas used here are standard, published methods. The implementation, interface, wording, findings logic and benchmark figures are original work by ITSM Compass. Results are estimates, not professional advice. See the terms of use.

© 2026 ITSM Compass. All rights reserved.

Questions

How much downtime does 99.9% availability allow?

On a 24×7 service, 99.9% availability allows about 8 hours 45 minutes of downtime a year, roughly 43 minutes 48 seconds a month or 10 minutes 5 seconds a week.

How do you calculate availability percentage?

Availability = (agreed service time − downtime) ÷ agreed service time × 100. Agreed service time is the hours the service is supposed to be available, such as 24×7 or business hours only, over the measurement window.

How do you calculate end-to-end availability across several systems?

If a service depends on components in series, multiply their availabilities: two components at 99.9% give about 99.8% end to end. If a component is duplicated so either copy can carry the load, its combined availability is 1 − (1 − A)².

Should an SLA use monthly or yearly availability?

Most SLAs measure monthly, because it matches reporting and service credits. Remember that a monthly 99.9% target allows only about 44 minutes of downtime in that month, so one bad incident can breach it.

SLAs look fine on paper but keep breaching?

An ITSM health check traces breaches back to the processes, suppliers and dependencies behind them.