All cases

Finding broken vending machines in hours, not days

A monitoring system now watches sales across De Bonte Koe’s vending-machine network and alerts the team when a machine unexpectedly goes quiet.

  • Monitoring
  • Automation
  • n8n
Client
Duration
2 weeks
Led by
Miloš
Category
Automation
A De Bonte Koe shop in a shopping centre, with a chocolate vending machine built into its orange front.~3 hto spot a quiet machine, down from 1–2 days

At a glance

What we did

  • Machine monitoring
  • Automated alerts
  • Testing

The challenge

A vending machine could stop working without anyone immediately noticing.

This had happened multiple times in practice: if nobody checked the machine or its sales data, it could remain offline for a day or two before the problem reached the right person.

How we helped

  1. 01

    Used sales activity as an early warning signal

    The system monitors payment activity across the vending-machine network and checks whether each machine is behaving as expected during its opening hours.

  2. 02

    Alerted the team when something looked wrong

    If a machine goes without sales for around three hours during its expected operating window, the responsible team members are notified automatically.

  3. 03

    Kept the issue visible until activity returned

    The system checks again every hour while the problem continues, then stops alerting once normal activity returns.

The outcome

Issues that had previously gone unnoticed for one or two days can now be surfaced in around three hours.

The system has already caught multiple real vending-machine problems, giving the team the chance to respond much earlier and reduce avoidable downtime.

Insights

  • Sales data doubles as a health check.

    Payment activity during opening hours shows whether each machine is working.

  • Alert the right people, at the right time.

    Only the responsible team members hear about it, and only after about three quiet hours.

  • Keep it visible until it’s fixed.

    Hourly reminders continue while the problem lasts, then stop on their own once sales return.

“Would highly recommend Milos. Fast learner, can operate in non structured/growing environment.”

Amy KleinChocolate Executive, De Bonte Koe