How to get notified the moment your site goes down

Most people find out their site is down the worst possible way: a customer emails to ask if something is broken. By then the outage has been running for who knows how long, and you are already starting from behind.

You can close that gap to a minute or two. The setup is genuinely quick, and the only real decisions are how often to check and how you want to be told.

Stop relying on yourself

Checking the site by hand does not scale, and it does not work at night. Browser extensions that ping a tab only run while your laptop is open. A cron job on the same server cannot tell you the server is down, because it went down too. Whatever watches your site has to live somewhere else and run on its own schedule.

That is what an external uptime monitor does. It sits outside your infrastructure and checks your public URL on a fixed interval, around the clock, whether or not anyone is awake.

The setup, start to finish

It really is just three steps:

1. Add your URL. The full address a visitor would type, including the https.
2. Pick a check interval. Every minute catches outages fast. Every five minutes is plenty for most sites and still means you hear about a problem long before your customers pile up.
3. Choose where alerts go. Email is the baseline and works everywhere. The important part is that the alert reaches you when you are not looking at the site.

From there the monitor does the relentless, boring work you would never keep up by hand: a check every interval, forever.

Get told when it recovers, too

A good alert has a matching all-clear. Knowing your site went down at 02:14 is half the story. Knowing it came back at 02:23 tells you the incident is over and how long it lasted. Those two timestamps are also the start of a useful habit: a record of every outage, so you can spot the ones that keep happening.

Avoiding alert fatigue

The fastest way to start ignoring alerts is to get too many of them. Two things keep them trustworthy:

• Confirm before alerting. A single failed check can be a momentary network hiccup. A monitor that re-checks before it emails you turns "ten panicked alerts a week" into "the two that were real".
• Alert on changes, not on every check. You want one message when the site goes down and one when it comes back, not a reminder every minute that it is still down.

Get those two right and your inbox stays quiet until it matters, which is exactly when you will actually pay attention.