Monthly Downtime Allowance by SLA Tier

Monthly Downtime Allowance Reference

This is a complete reference table showing the maximum allowed monthly downtime for every commonly used SLA percentage. The calculations assume a 30.44-day average month (365.25 / 12).

Complete Monthly Downtime Table

SLA %	Nines	Monthly Downtime	In Minutes
95%	—	1 day, 12 hours, 29 min	2,191.5
99%	Two	7 hours, 18 min, 18 sec	438.3
99.5%	Two and a half	3 hours, 39 min, 9 sec	219.2
99.8%	—	1 hour, 27 min, 39 sec	87.7
99.9%	Three	43 min, 50 sec	43.8
99.95%	Three and a half	21 min, 55 sec	21.9
99.99%	Four	4 min, 23 sec	4.4
99.995%	Four and a half	2 min, 12 sec	2.2
99.999%	Five	26 sec	0.44
99.9999%	Six	2.6 sec	0.044

Scaling by Time Period

You can derive other time periods from the monthly figure:

To get...	Multiply monthly by...
Yearly downtime	12
Weekly downtime	0.23 (÷ 4.35)
Daily downtime	0.033 (÷ 30.44)

Practical Interpretation

• 7+ hours/month (99%): Room for weekly 1-hour maintenance windows
• 43 minutes/month (99.9%): Room for one ~30-minute incident or a few short ones
• 4.4 minutes/month (99.99%): A single moderate incident can exceed the budget
• 26 seconds/month (99.999%): Essentially requires zero human-caused downtime

The Half-Nine Steps

Notice the "half nines" (99.5%, 99.95%, 99.995%) — these are useful intermediate targets:

• 99.5% → 99.9%: Halves the downtime (from 219 to 43 minutes)
• 99.95%: A popular target that is more achievable than 99.99% while being noticeably better than 99.9%
• 99.995%: Rarely used — teams usually jump from 99.99% to 99.999%