Uptime Calculator
An availability percentage is an outage allowance written in a way that hides how big it is. Enter the number from the SLA and get the downtime it actually permits per day, week, month, quarter and year — plus the reverse, for working out what last week's incident cost you.
Free · No signup · Runs entirely in your browser
An SLA percentage is an allowance in disguise. This uptime calculator converts the number in the contract into the downtime it actually permits, over each period anyone measures. Figures follow uptime.is: a Gregorian year, a month of exactly a twelfth of it.
The percentage in the SLA. Try 99.95 or 99.99.
| Period | Allowed downtime |
|---|---|
| Daily | 1m 26.4s |
| Weekly | 10m 4.8s |
| Monthly | 43m 49.7s |
| Quarterly | 2h 11m 29.2s |
| Yearly | 8h 45m 57s |
After an incident
That is 99.8973% availability — outside a 99.9% commitment.
| Uptime | Daily | Weekly | Monthly | Yearly |
|---|---|---|---|---|
| 99% | 14m 24s | 1h 40m 48s | 7h 18m 17.5s | 3d 15h 39m 29.5s |
| 99.5% | 7m 12s | 50m 24s | 3h 39m 8.7s | 1d 19h 49m 44.8s |
| 99.9% | 1m 26.4s | 10m 4.8s | 43m 49.7s | 8h 45m 57s |
| 99.95% | 43.2s | 5m 2.4s | 21m 54.9s | 4h 22m 58.5s |
| 99.99% | 8.6s | 1m 0.5s | 4m 23s | 52m 35.7s |
| 99.999% | 0.9s | 6s | 26.3s | 5m 15.6s |
An availability target is a promise the architecture has to keep. If you are building the system largely on your own, Tekk turns what you want into specs your coding agent can actually execute.
The formula
allowed downtime = period length × (1 − uptime)
That is all of it. 99.9% means a thousandth of the period may be spent offline, so a month of 30.4369 days permits 30.4369 × 1440 × 0.001 = 43.8 minutes.
The standard tiers
| Uptime | Daily | Weekly | Monthly | Yearly |
|---|---|---|---|---|
| 99% | 14m 24s | 1h 40m 48s | 7h 18m 17.5s | 3d 15h 39m 29.5s |
| 99.5% | 7m 12s | 50m 24s | 3h 39m 8.7s | 1d 19h 49m 44.8s |
| 99.9% | 1m 26.4s | 10m 4.8s | 43m 49.7s | 8h 45m 57s |
| 99.95% | 43.2s | 5m 2.4s | 21m 54.9s | 4h 22m 58.5s |
| 99.99% | 8.6s | 1m 0.5s | 4m 23s | 52m 35.7s |
| 99.999% | 0.9s | 6s | 26.3s | 5m 15.6s |
Each nine divides the allowance by ten. That is the fact the percentages disguise — 99.9% and 99.99% look like neighbours and are an order of magnitude apart.
Why the month convention matters
There is no agreed length for "a month", and the choice is visible in the output:
| Month defined as | Downtime at 99.9% |
|---|---|
| 30 days | 43m 12s |
| 31 days | 44m 38.4s |
| A twelfth of a Gregorian year (30.4369 days) | 43m 49.7s |
The figures on this page use the last one, matching uptime.is. The differences are small enough to ignore in planning and large enough to matter in a dispute, so if you are writing an SLA, say which one you mean.
Reading the number honestly
Three things to hold onto.
Frequency is not in the number. Eight hours down once a year and ninety seconds down every day are both 99.9%. Users experience these very differently, and only one of them costs you the account.
Measurement is contested. Uptime measured by your own health check from inside your own network is not the same as availability measured by a customer's request from another continent. Whose measurement counts belongs in the contract.
Credits are not compensation. A typical SLA refunds a slice of the monthly fee, which bears no relationship to what the outage cost the customer. Treat the percentage as a statement of engineering intent, not as insurance.
How it works
- 1
Enter the percentage from the contract
99.9%, 99.95%, 99.99% — whichever tier you are signing or selling. The uptime calculator multiplies the period by one minus that figure, which is the entire arithmetic. Everything difficult about this is in the period conventions, not the maths.
- 2
Read the allowance per period
Any uptime percentage calculator will give you one period; the point of showing five is that the same commitment looks harmless daily and alarming annually. Three nines is 1m 26s a day, which nobody would notice, and 8h 45m 57s a year, which is a full working day of outage. Quoting the daily figure and budgeting for the annual one is how availability targets get agreed too casually.
- 3
Work backwards after an incident
The panel below the table takes an outage length and a measurement window and returns the availability you actually delivered. That is the direction you need when something has already broken and someone is asking whether the SLA uptime calculator says you owe a credit.
Frequently asked questions
- How much downtime is 99.9% uptime?
- 1m 26.4s a day, 10m 4.8s a week, 43m 49.7s a month, and 8h 45m 57s a year. The annual figure is the one worth quoting in a planning conversation — three nines permits most of a working day offline.
- How much downtime is 99.99% uptime?
- 8.6s a day, 4m 23s a month and 52m 35.7s a year. Four nines is the point where a single bad deploy can consume the entire annual budget, which is why it usually implies automated rollback rather than a person being paged.
- What is the formula for uptime?
- Allowed downtime equals the length of the period multiplied by one minus the uptime fraction. For 99.9% over a 30.44-day month: 30.44 × 24 × 60 × 0.001 = 43.8 minutes. This uptime percentage calculator runs that against every period at once so you do not have to pick one first.
- How many days does this assume in a month?
- 30.4369 — a twelfth of a Gregorian year of 365.2425 days. This matters more than it sounds: a 30-day month gives 43m 12s at three nines while a twelfth-of-a-year month gives 43m 49.7s. We follow the convention uptime.is uses, because that is the reference most SLA discussions end up quoting.
- Does an SLA uptime calculator account for maintenance windows?
- No, and neither does this one. Most commercial SLAs exclude scheduled maintenance announced in advance, so the contractual number and the number your monitoring reports are often different. An SLA uptime calculator gives you the raw allowance; the exclusions are a contract-reading exercise.
- What uptime should I promise?
- Less than you think you can hit, because the penalty is asymmetric — exceeding your commitment earns nothing, missing it costs credits and trust. Most SaaS lands on 99.9% for standard plans and 99.95% or higher for enterprise tiers, priced accordingly.
- What is the difference between availability and reliability?
- Availability is the share of time the system responds; reliability is how often it fails. A service that goes down for one second every hour has excellent availability and terrible reliability. Users usually notice the second, which is why frequency and duration are worth tracking separately.
- Why is each extra nine so much harder?
- Because each one cuts the allowance by 90%. Going from 99.9% to 99.99% takes your monthly budget from 43m 49.7s to 4m 23s — less time than most teams need to notice an alert, decide it is real, and get to a keyboard. Past three nines, the constraint stops being effort and becomes architecture.
- What does the reverse mode do?
- It takes the length of an outage and the window it happened in, then returns the availability you delivered. A 45-minute outage in a month is 99.9%, so it just misses a three-nines commitment. Working it out this way is what an uptime calculator is actually used for once something has broken.
- Do SLA credits make up for the downtime?
- Almost never. A typical credit refunds a percentage of the monthly fee for the affected service, which is unrelated to what the outage cost the customer's business. Credits are a signalling mechanism, not compensation, and buyers who treat them as insurance are usually disappointed.
- Can I use this for a percentage between the standard tiers?
- Yes. The field takes any value between 0 and 100, so this uptime calculator handles 99.92% or 99.5% the same way as the round numbers. The reference table at the bottom covers the standard tiers for when you are choosing one rather than checking one.
- Do you store the numbers I enter?
- No. The calculation runs entirely in your browser. Nothing is sent to a server, nothing is logged, and there is no account. Disconnect from the internet and it will still work.
- Why is this free, and what is Tekk?
- Tekk is a spec-driven development platform for people building software with AI coding agents. This calculator costs us nothing to run, and the people writing availability targets are often the same people building the system that has to hold them. No signup, no run limit, no upsell inside the tool.
Want a real spec for what you’re building?
Drop a sentence. Tekk grounds it in your actual code and turns it into an executable plan.
Free to try · Connect GitHub during signup