What Is Website Uptime and Why Does It Matter?
Uptime is the share of time your website is available to visitors. Here's how it's calculated, what the "nines" really mean in minutes of downtime, and why measuring it yourself matters.
Table of contents
Every website owner eventually asks the same question: is my site actually online when people try to visit it? Uptime is the number that answers it. It sounds simple, but the way uptime is measured, reported and promised has a lot of nuance. This guide explains what uptime is, how it's calculated, what common targets like 99.9% really mean, and how to start measuring your own.
What does "uptime" mean?
Uptime is the proportion of time a website or service is available and working as expected. Its opposite, downtime, is the time when visitors cannot use it, whether because the server is unreachable, the page returns an error, or the connection fails.
Uptime is almost always expressed as a percentage over a period, for example "99.95% uptime over the last 30 days". A website that was never unavailable during that period has 100% uptime.
The important word in the definition is available. A server can be powered on and still be useless to visitors: the web server process may have crashed, the database may be unreachable so every page shows an error, or the SSL certificate may have expired so browsers refuse to load the page. Good uptime measurement looks at the website from the outside, the way a visitor would, rather than asking whether a machine is switched on.
How is uptime calculated?
The basic formula is straightforward:
uptime % = (time available ÷ total time) × 100In practice, nobody watches a website continuously. Monitoring services check it at regular intervals and treat each check as a sample. If a website is checked every 5 minutes, it receives 288 checks per day. If 286 of them succeed, the measured uptime for that day is:
286 ÷ 288 × 100 = 99.31%This sampling approach has two consequences worth understanding:
- The check interval limits precision. With 5-minute checks, an outage of 90 seconds that happens between two checks may never be detected. Shorter intervals catch shorter outages.
- Not every failed check is an outage. A single failed request can be caused by a brief network problem somewhere between the monitoring server and your website. That's why good monitoring tools confirm a failure with a second check before declaring the site down.
What do the "nines" mean in real time?
Uptime targets are often described in "nines": two nines is 99%, three nines is 99.9%, and so on. The percentages look almost identical, but the allowed downtime differs enormously. Here is the maximum downtime each level allows, calculated from the number of minutes in each period:
| Uptime | Per day | Per 30-day month | Per year (365 days) |
|---|---|---|---|
| 99% | 14 min 24 s | 7 h 12 min | 3 days 15 h 36 min |
| 99.5% | 7 min 12 s | 3 h 36 min | 1 day 19 h 48 min |
| 99.9% | 1 min 26 s | 43 min 12 s | 8 h 45 min 36 s |
| 99.95% | 43 s | 21 min 36 s | 4 h 22 min 48 s |
| 99.99% | 8.6 s | 4 min 19 s | 52 min 34 s |
Seen this way, "99% uptime" is not impressive at all: it allows more than seven hours of downtime every month. For most business websites, 99.9% is a reasonable baseline, while 99.99% usually requires redundant infrastructure across multiple servers or regions.
Why uptime matters
Visitors don't wait
When a page fails to load, most people don't try again later. They return to the search results or open a competitor's website. For an online store, a booking system or a lead-generation page, every minute of downtime is a minute in which nobody can buy, book or contact you.
Trust is hard to rebuild
A customer who sees an error page once may forgive it. A customer who sees it repeatedly starts to question whether the business behind the website is reliable. Downtime during a campaign, a product launch or a busy season is especially costly because that's when the most people are paying attention.
Search engines notice
Search engine crawlers visit websites regularly. If they repeatedly receive server errors or timeouts, they may reduce how often they crawl, and long or frequent outages can affect how pages are indexed. Short, rare outages are normal on the web and are not a reason to panic, but persistent availability problems are worth fixing for SEO reasons alone.
You can't fix what you don't know about
The most practical reason to care about uptime is simple: without measurement, you discover outages from customers, colleagues or your own accidental visit, often hours after they started. Measuring uptime turns a vague worry into a concrete number and gives you an early warning when something breaks.
Uptime vs. availability vs. SLA
These terms are related but not identical:
- Uptime usually refers to the measured percentage of time a service was working.
- Availability is often used as a synonym, though in engineering it can include whether the service performed acceptably, not just whether it responded.
- An SLA (service level agreement) is a contractual promise, for example from a hosting provider, that a service will reach a certain uptime, often with compensation if it doesn't.
A hosting provider's SLA is useful, but it measures their infrastructure. Your website can still be down while the host meets its SLA, for example because of a broken plugin, a full disk, an expired domain or a misconfigured DNS record. That's why independent monitoring of the website itself is valuable.
What causes downtime?
Outages rarely have a single cause. Common categories include server and hosting failures, software errors after updates, traffic spikes that exhaust resources, DNS problems, expired SSL certificates and expired domain registrations. We cover these in detail in What Causes Website Downtime?
How to measure your website's uptime
Measuring uptime yourself is easier than it sounds. You need a service that:
- requests your website from outside your own network at a regular interval;
- records whether each request succeeded, the HTTP status code and the response time;
- confirms failures before treating them as outages;
- alerts you when an outage is confirmed and when it ends;
- keeps a history so you can calculate uptime over days and weeks.
This is called uptime monitoring, and our complete guide covers it in depth. It's exactly what UptimeStats does. The Free plan checks up to 5 websites every 5 minutes, confirms failures with a second check, emails you once when an outage is confirmed and once when it recovers, and shows uptime for the last 24 hours, 7 days and 30 days. If you want to learn the fundamentals first, read our beginner's guide to website monitoring.
For a one-off check right now, try the free website status checker, or read how to check if a website is online for all the methods.
Key takeaways
- Uptime is the percentage of time your website is available to visitors, measured from the outside.
- Small differences in percentages hide large differences in downtime: 99% allows over 7 hours per month, 99.9% about 43 minutes.
- Monitoring measures uptime by sampling at intervals, so the check frequency and outage confirmation matter.
- Hosting SLAs don't replace monitoring your own website, because many outages happen above the infrastructure level.
