Website uptime explained: what 99.9% really means

Uptime measures how often your website can be reached. When a visitor tries to load a page and the server answers, that counts as up. When the server does not answer, or answers with an error, that counts as down.

The figure is usually a target for the hosting service, covering the servers and the network. It does not cover problems inside your own site, such as a broken plugin or a database error, which can make a site fail on a perfectly healthy server.

Planned maintenance is often excluded from the figure, so read how a provider defines it.

Two hosts can both claim 99.9% and behave very differently. One may have a single long outage and the other many short ones. One may tell you quickly what is happening and the other may leave you guessing.

When you evaluate a host, ask how it communicates during incidents, how quickly support responds and how it prevents repeat problems. A status page, clear notices and fast answers are signs of a provider that takes reliability seriously.

Your own preparation matters too. A recent backup and a clear idea of who to call turns a stressful outage into a short one.

Monitoring tools request a page from your site every minute or few minutes, from one or more locations. If the request succeeds, the site is up. If it fails or times out, it is down. Uptime is the percentage of checks that succeeded over a period.

Different tools check from different places and at different intervals, so their numbers can differ slightly. A tool that checks every five minutes can miss a short outage entirely, and one that checks every minute will catch it.

If you want a reliable independent record, use a monitoring service and keep its history. It is useful evidence if you ever need to raise a complaint, and a useful way to see trends.

If your site goes down, write down the time it started, what you saw, what you did and when it recovered. Over a year, this short log shows patterns, such as outages after updates or at particular times, and gives you clear facts to share with support.

It also helps your own planning. If most trouble follows plugin updates, you know to test them on a copy first. If it follows traffic spikes, you know when to upgrade. A few lines per incident is all it takes.