From Smart Data to AI
The technology changes. The operating question stays remarkably consistent.
Technology changes and evolves. The executive job remains: understand, structure, decide, and turn it into operating models that create value.
2019: make data operational
At A.T.U. we combined internal signals such as web, CRM and workshop data with external context including geography, advertising, competition, seasonality and weather. The purpose was not “more data”. It was better allocation: where should marketing spend go, for which customer, at what moment, given actual operating capacity?
One rule made the difference: no free capacity in a workshop, no ad for it. Media spend followed real capacity, so the customer who clicked also found an appointment. Weather did the same job from the outside: hail in Munich, message out.
2026: same question, new interface
AI changes the speed and accessibility of analysis, synthesis and knowledge work. But the management problem remains familiar: What context matters? Which assumptions are weak? Which constraint is real? Who owns the decision? What changes the economics?
From information to operating model
That is why I see AI as the next chapter in a longer sequence: GSM payments, mobile software, M2M, e-commerce, SaaS, data and now AI. The technology is not the destination. The value appears when it changes how a business operates.
The test I still use is a question, not a dataset: what decision does this data make possible?