free_tool
SLO & Error Budget Calculator
"Five nines" sounds reassuring until you see it in minutes. Pick an availability target and get the downtime it actually permits, a request error budget, and the burn rate that says when to stop shipping features and go fix reliability.
Allowed downtime at 99.9%
43m 12s/ 30d
1m 26s / day · 8h 46m / year
2.5K of 5K failures still in budget.
Want these numbers to mean something in production? I'll help you set SLOs that map to real user pain, then wire the alerts to enforce them.
Set up SLOs that hold: book a callBudgets shown over a 30-day window. Pure arithmetic from the target, with no hidden assumptions. Share the link to compare targets with your team.
how_it_works
From a target to an enforceable budget
Your error budget is simply the gap between your target and 100%. A 99.9% target means 0.1% of requests, or about 43 minutes a month, are allowed to fail. That budget is something to spend deliberately on risky deploys, not a number to fear.
The burn rate tells you how fast you're using it. Burning at 1× means you'll spend exactly the month's budget over the month; at 14.4× you'll blow a 30-day budget in roughly two days, the classic threshold for a fast-burn page. When the budget's gone, the policy writes itself: stop shipping, fix reliability.
faq
Questions & answers
- How does the SLO calculator turn an availability target into downtime?
- It multiplies the unavailable fraction, one minus your target over 100, by the length of each window. A 99.9% target leaves about 43 minutes of allowed downtime per 30-day month, and it also shows the figure per day, week, quarter and year.
- What is the error budget in requests?
- It is the share of requests you are allowed to fail over the month: the unavailable fraction times your monthly request count. At 99.9% over 5 million requests, that is 5,000 failed requests before the budget is spent.
- What does the burn rate row mean?
- It shows how fast you would exhaust a 30-day budget at 1x, 2x, 5x, 10x and 14.4x consumption. The 14.4x figure is Google's fast-burn alert threshold, the point where the monthly budget would be gone in about two days.
- Does this tool read my real availability data?
- No. You type in the numbers and every calculation happens in your browser, with nothing sent to a server. You can compare a measured availability you enter against the budget, but the tool does not pull live metrics.
- Does it handle SLOs across multiple dependent services?
- No. It models a single SLO over a 30-day window and does not compose availability across services or account for cascading failures, so chain those numbers yourself if a request spans several systems.
Want these numbers pressure-tested on your stack?
I'll review your inputs and tell you where the real cost and risk are. Book a call, or leave your email and I'll reach out.
Prefer proof first? See how this plays out in real case studies →