layman/terms
Going live

Monitoring

mon · ih · tur · ing

In shortMonitoring is software that keeps watching a website or app day and night and raises the alarm the moment something goes wrong.

Nobody stays awake all night watching a website. Monitoring does it for them.

A smoke alarm doesn’t sleep. It sits on the ceiling quietly sniffing the air. The moment there’s smoke it beeps so you can deal with it before it’s a fire.

Monitoring is that smoke alarm for a website or app. It keeps checking that everything works and raises the alarm as soon as something doesn’t.

What it watches

Why it matters

Without it, the first person to notice a problem is a customer who usually just leaves. With it, the team often fixes things before most people notice anything at all.

What it isn’t

It doesn’t fix anything. It only tells someone. And it isn’t the same as a log: logs are the diary of what happened, monitoring is the alarm that says “look now”.

Where the comparison breaks down

A smoke alarm watches for one thing. Monitoring can watch hundreds at once and draw graphs of them. And instead of beeping in the kitchen, it sends a message to a phone, often to whoever is on duty that night.

How the words fit together

Monitoring tracks uptime and spots outages. When it raises the alarm, people read the logs to find out why. Error status codes and a struggling server are two of the things it watches for.

Pro tip

Only alert people for things they need to act on. If the alarm goes off every hour, people learn to ignore it, just like a smoke alarm that beeps every time you make toast.

When you'll hear it

“I got paged at 3 am.” The monitoring spotted a problem and woke up whoever was on duty.

“Is it on the dashboard?” Can we see it on the screen that shows how the system is doing?

“We found out from a customer.” The monitoring missed it, which is the worst way to find out.

Say it in a meeting

“If the site went down at 2 am, how long before someone knew?”