We very often think in terms of perfect target states and automatically have the 100 percent result in mind. The problem is: those who want to do everything perfectly often never start at all.

Yes, absolutely. The admittedly stereotypical image of German companies is strongly shaped by perfection and a pronounced zero-error tolerance. This mindset has its justified origins in manufacturing, but over the years it has been transferred to entire organizations.

At the same time, it’s exactly what made us famous: Quality made in Germany.

But the world has changed. Market pressure, speed, and customer expectations work very differently today, even in our private lives. If I can get something faster and cheaper, I’m quite willing to accept compromises in perfection.

One defining experience for me was in workshops with American colleagues. After I presented a very well-thought-out vision aimed at a perfect target state, they became visibly nervous. They wanted to move straight into execution, even though not all details had been clarified. It shows: in other cultures, 80 percent is often good enough.

The first step was establishing a genuine culture of learning from mistakes, simultaneously “at the top” and “at the bottom.”

In leadership meetings, after an incident, the question often came up very quickly: Who’s going to take the fall for this? That shows how human the fear of mistakes is. But this kind of leadership creates insecurity and blocks innovation.

For teams, it simply felt safer to work for a long time on a spectacular target state rather than move into execution.

After incidents, I consistently introduced lessons-learned formats and at the same time stopped looking for people to blame. I’m interested exclusively in the cause and in how we can fix it quickly. When a system outage occurs, my approach is:

Acknowledge the error, focus on the solution: technically, professionally, and organizationally. Then sleep on it for a night and only afterwards conduct a clean root-cause analysis.

And very importantly: we celebrate finding errors quickly.

They say you have to say something reliably seven times. But what really matters is doing it reliably seven times. Only when the organization experiences in a real incident that you keep your word and don’t put anyone on display does trust emerge.

Realistically: about two years.

By people having the courage to admit them.

Our average system outage duration has dropped from 4–8 hours to today 30 to a maximum of 120 minutes.

At first, I accept that as well. Stay calm, fix the error, which is faster if we already know it and then analyse the root cause.

A third occurrence of the same error is extremely rare. If it does happen, we have a personal conversation. Because we need people who think along with us.

Yes, but never in front of the entire team. Always one-on-one and in a reasonable, respectful atmosphere.

The team itself describes a mindset of: We fix errors together.

I notice it in the fact that IT no longer hesitates to introduce new technologies. We’ve become much faster at delivering new solutions.

We still plan, but no longer only for perfection. We finally have the courage to try new things.

I report directly to the board or executive management. At the beginning, I always addressed the foundations of how we work together.

For me, two points are central: maximum commitment and 100 percent transparency. When someone asks for a culprit, I consistently stay with myself as the person responsible for IT as a whole.

That consistency has an effect: after some time, the question (about whom to blame) simply stops being asked.

I believe so. The board gets someone who takes responsibility instead of looking for scapegoats. And they see that it works: We’ve reduced our total system outages from over 48 hours per year to under 10 hours.

Because it’s always worth pursuing ambitious goals, growing through them, and celebrating successes together.