SLA Uptime Calculator
What 99.9% really allows: downtime per month, SLA credit math, composite SLAs.
of downtime allowed per month at 99.9%. You actually hit 99.8973% with 45 min down — that breaches the 99.9% tier, an estimated $500 credit (10% of fee).
| Period | Allowed downtime at 99.9% |
|---|---|
| per day | 1.4m |
| per week | 10.1m |
| per month | 43.8m |
| per year | 8.8h |
Credit tiers modeled on common cloud SLAs: <99.9% → 10%, <99% → 25%, <95% → 100% of monthly fee. Check your contract — every vendor differs.
How it works
Availability percentages are deceptively compact. The difference between 99.9% and 99.99% reads like a rounding error on a contract page, yet it is the difference between 43.8 minutes and 4.4 minutes of allowed downtime each month — a tenfold change in how your team must engineer, staff and alert. This calculator converts any nines target into concrete downtime budgets per day, week, month and year, so the contract language turns into numbers an on-call rotation can actually plan around. Everything computes locally in your browser; nothing you enter is sent anywhere.
The composite section handles the math most teams skip. When your service calls dependencies in series — an AI provider, a database, an auth vendor — the effective availability of the whole path is the product of each link. Chain three services at 99.9% and the best you can honestly promise is 99.7%, more than two hours of monthly exposure. Add each dependency in your critical path and the tool shows the effective ceiling before you write an SLA your architecture cannot keep. The fix is either fewer serial dependencies, graceful degradation when one fails, or a more honest number in your own contract.
The credit estimator applies the tier structure most cloud vendors use — roughly 10% of the monthly fee below 99.9%, 25% below 99%, and a full refund below 95% — to your actual fee. The output is usually humbling: a serious outage on a four-figure contract returns a three-figure credit. That asymmetry is worth knowing before you build a vendor strategy around penalty clauses rather than redundancy.
Finally, the inverse mode grades a month you already lived through. Enter the downtime minutes from your incident log and get the achieved percentage, the nines tier you actually hit, and the credit you could claim. Pair it with the incident postmortem generator after an outage, and with provider status history when deciding which vendors deserve a place in your critical path at all.
Frequently asked questions
How much downtime does 99.9% availability actually allow?
About 43.8 minutes per month, 10.1 minutes per week and 8.8 hours per year. The jump between tiers is brutal: 99.99% allows only 4.4 minutes a month, and 99.999% just 26 seconds. Each extra nine cuts the budget by a factor of ten, which is why every additional nine costs disproportionately more in redundancy, on-call coverage and engineering time than the one before it.
How do composite SLAs work when my service depends on several vendors?
If your request path goes through services in series, their availabilities multiply. Two dependencies at 99.9% each give you at most 99.8% — roughly 87 minutes of monthly downtime budget, double what either vendor promises alone. This is the most common SLA planning mistake: teams promise customers the same nines their cheapest dependency offers, forgetting that every additional link in the chain subtracts from the budget.
Are SLA credits actually worth pursuing?
Usually not financially — a 10% credit on a $5,000 monthly bill is $500, far less than the engineering time spent documenting the outage and filing the claim. Credits matter more as negotiation leverage at renewal and as a forcing function: vendors that pay credits track availability honestly. Treat the credit math here as a way to quantify what an SLA promise is really worth, not as a revenue stream.
What is the difference between SLA, SLO and SLI?
An SLI is the measured indicator, such as the fraction of successful requests. An SLO is the internal objective your team commits to, for example 99.95% over 30 days. An SLA is the external contract with financial consequences when breached. Healthy teams set SLOs stricter than their SLAs so internal alarms fire well before contractual penalties do, leaving room to respond before customers are owed money.
How do I use the inverse mode to grade a month we already had?
Enter the total minutes of downtime you recorded this month and the calculator converts it to an achieved availability percentage, then tells you which nines tier you actually landed in and whether common credit thresholds were breached. This is useful for postmortems and quarterly vendor reviews: instead of arguing about feelings, you can state that 45 minutes down equals 99.897%, just below the three-nines line.
Built by FORG — AI cost observability for agentic coding. Free tools, no signup, nothing leaves your browser.
Learn about FORG