Advice
Five signs nobody uses your dashboard
It looks spectacular at the kickoff and is dead three months later. The symptoms are always the same.
The five signs
- Nobody can say what action they take when the number turns red. A metric with no attached action is decoration.
- It refreshes slower than the decision it supports. Daily data cannot drive a per-shift decision.
- It needs an expert to be read. If you have to explain the chart, the chart is wrong.
- People still export to Excel next to it. That export is the real specification of what they needed.
- Whoever asked for it is not whoever reads it. The most expensive mistake, and the most common.
The fix
The fix is rarely visual. It is sitting down again with the decision-maker, understanding their real cycle, and cutting the dashboard down to what actually changes behaviour.
In practice that usually means deleting two thirds of the charts. Nobody misses any of them.
“If you switch the dashboard off and nobody calls within a week, it was not a dashboard.”
Blog
Advice
Your problem is not AI — nobody trusts the data
Before talking predictive models, there are three questions almost no plant can answer about its own data.
Architecture
MQTT Sparkplug B: why I stopped writing point-to-point integrations
Six plants, four protocols and a separate integration per pair. The predictable ending, and how I reversed it.