How long is a system down at your company before anyone notices?

How long is a system down at your company before anyone notices? Two minutes are enough.

Without an alert it goes like this: the server with orders, mail or filing fails, and it is noticed only when somebody needs something and gets nothing. At the weekend or during the holidays that can take hours. In the worst case the customer notices first.

With an alert the order is reversed. At our company a second machine asks every minute — not whether the server is switched on, but whether it is really working. If the answer fails twice, the phone rings. While the fault lasts, a reminder follows every 30 minutes. When the server runs again, the all-clear arrives with the duration of the outage.

An AI-run back office also works at night and at the weekend: it sorts mail, files documents, prepares replies. That is exactly why this alert belongs to it. Whoever hands work to machines has to be the first to know when one stands still.

How an alert has to be built

It runs on a second device — not on the machine it watches. Otherwise it goes down with it and stays silent exactly when it should speak.

It does not merely check whether the machine is «there», but whether it answers the way it does in operation. The difference matters: a machine can still respond to a simple knock on the network and yet no longer accept a single new connection. A watchdog that only knocks sleeps through exactly that outage.

And it reaches the phone by a path of its own that does not lead through the watched server. Otherwise the message gets stuck exactly where the fault is.

Two questions for every company with its own technology

Who finds out, and how fast? A notification on the phone, not a red light in a room where nobody is.
Who acts? A name and a deputy, before the holidays begin. Software cannot answer this question.

Operating instead of owning

Your own systems pay off because the data stays in the house. But they come with a task that appears on no invoice: somebody has to operate them. For small companies that is often the reason not to try at all. This is exactly the part we take over: Your back office, run by AI.