A good 35 years ago, "digitalization in manufacturing" was not a buzzword. It was more of an improvisation project with a serious background: quality and safety became more demanding - at the same time, materials became lighter, tolerances tighter and processes more sensitive.
The problem back then was that the tools were still "blind". Hardly any sensors, hardly any data. But the expectation was already: precise, reproducible, traceable.
So we took a detour that, looking back, seems like the first real step towards Industry 4.0:
👉 We built intelligent testing equipment first.
Not because it was "cool" - but because it was necessary. If the production tool has no sensors, then you need something that can measure whether it is still working within its limits. We were only able to qualify "primitive" tools by using test equipment with sensors.
Later, the tools became smarter: displays, controllers, more signals, more data. And suddenly we had numbers everywhere - which initially felt like progress.
But in my view, the real breakthrough came with another realization:
Data is only valuable if it is authentic.
And authenticity rarely arises when a manufacturer "proves" its own hardware - or when a producer programs its own truths.
The key was independent software:
Systems that collect and evaluate data independently of hardware manufacturers - and thus create a neutral basis that can be relied upon. In some areas, literally: on which you can trust your life.
Today, there is a renewed sense of momentum in IT. Many software companies are able to pick up on trends more quickly and bring real added value to production lines.
At the same time, however, I am also observing a counter-movement:
More and more manufacturing companies are trying to build everything themselves - from cars to their own ERP or data platform.
And I totally understand the impulse. Control, speed, independence.
However, without focus and experience, this can quickly backfire.
The result is data silos, redundancies, different truths, unstable systems - and ultimately decisions based on data that nobody really trusts anymore.
And as if that wasn't enough, the complexity of regulation (especially in the EU) is growing in parallel:
Secure products are no longer enough. Software must also be secure, traceable, compliant and resilient against attacks - but we see the opposite every day.
My takeaway after all these years:
Digitalization is not a question of "more data".
It's about trustworthy data, clean architecture and clear accountability.
Perhaps that is the most important lesson from 35 years of manufacturing:
Technology only brings progress if it is credible.
How do you see it - where are you currently experiencing "build vs. buy" in production? And how do you decide whether data is really reliable?