Get Ready to Fail

The Comfortable Trap

Many medium-sized companies in Germany are world market leaders in their field or are praised as hidden champions. With diligence and technical competence, Germany stands as a robust industrial location with high quality standards. We want to maintain this status. Traditional companies repeatedly go through phases in which they reflect on their "core competencies." We identify strengths and weaknesses to strategically position the organizations from within. In this process, it is not uncommon for considerable discrepancies to arise with market conditions and customer requirements. In addition, a certain complacency develops. Existing competencies are merely used but not necessarily further developed. Sometimes we lack the courage to make radical changes. In Germany in particular, solidity stands for a fundamental value. Traditional companies make decisions that are safe. A good business model must be profitable as quickly as possible.

The Heat is on

There is a great danger that we will run out of time. New companies are coming onto the market with new technologies that have the potential to destroy existing structures — disruption is everywhere. There is always the temptation to hope that challengers from outside will stumble. BYD was written off for years, until it became the world's largest electric vehicle manufacturer. Amazon was considered a bookseller. These stories are now history. The more relevant question today is what comes next — and how fast.

What has changed since the emergence of large language models and generative AI is the pace of that question. AI is no longer a topic for the innovation department. It sits at the top of every CEO and CIO agenda, driven by the realization that these capabilities are already reshaping engineering, production planning, after sales, and supply chain management in parallel. According to a 2026 survey of 300 manufacturing professionals, 98% of manufacturers are exploring or actively considering AI-driven automation — yet only 20% feel fully prepared to deploy it at scale. That gap between awareness and readiness is exactly where disruption enters.

The transition now underway is not from manual to digital. It is from AI as an assistant to AI as an agent — systems that do not merely surface insights but execute decisions autonomously, coordinate across workflows, and adapt in real time. Companies who are still running pilots and proofs of concept are no longer in an early position. They are falling behind organizations that have moved to operational deployment.

Disruptive Thinking

Disruptive business models are not created for the sake of destroying other companies. Rather, it is about answering the basic question: What can I do to create customer value? And: What technology can I use to achieve this as efficiently and effectively as possible? It is precisely this approach that poses major problems for established companies. The company's successfully elaborated core competencies are suddenly no longer relevant for solving the identified customer problems. To think disruptively, we cannot continue to do things from within and in an evolutionary manner but must drastically reprogram our behavior.

Consider what is happening at the intersection of manufacturing and service. Machine builders who once sold equipment are now moving toward Equipment-as-a-Service models, where the customer pays for uptime rather than assets and the manufacturer retains responsibility for performance. This is not a technology story. It is a customer value story, enabled by technology. The companies pursuing it are not doing so because they love disruption. They are doing so because their customers have stopped wanting to own complexity.

Impetus from Outside

Radical impulses rarely come from within. Many companies have already recognized this. Clinging to the compulsion to succeed means that new models that go beyond the limits of our own competence are not considered or tried out. We want to believe too much in supposedly realistic and feasible concepts. However, if we ask too early about the prospects of success, innovative concepts will not be addressed or tried out.

The tendency to evaluate new ideas against the standards of the existing business is the most common way that established organizations neutralize genuine innovation before it has a chance to prove itself. An idea that threatens the current model will always look financially unattractive when measured by the current model's criteria. That is not analysis. That is self-protection dressed as prudence.

The Force of Change

Meanwhile, most companies are taking startups and ideas from other sectors seriously and looking beyond the horizon of their own area. But we need to change our behavior much more if we want to keep up with the force of new business models from other regions of the world. An established corporation must think very carefully about how strongly a team that is to work on new ideas and models is integrated into the corporate structures. The group's top management must actively participate in the development of new models to recognize what strategic implications they may have for the basic business. And this should be done without putting on the brakes for fear of inner danger at the same time.

This is particularly urgent now. Agentic AI is moving into real operational environments. On the factory floor, an AI agent no longer just predicts equipment failures — it ingests sensor data, production schedules, and maintenance history to draft a specific repair plan and schedule it without waiting for a human to notice the problem first. Those capabilities compound. An organization that starts building them today will be structurally more capable in two years than one that waits for a clearer picture. The learning rate of the organization has become a competitive moat.

Free Space

Financial and time freedom must be created in such a way that things can be developed, tried out and, if necessary, scrapped again. Shareholders do not allow established groups to incur losses. But these must be accepted to spin off new ideas. Perhaps there is an idea that will ensure the company's success in such rapidly changing times. And if it goes wrong, you take the experience with you for the next idea.

The cost of running experiments with AI has dropped dramatically. What required a dedicated data science team and months of development two years ago can now be prototyped in days. The barrier to trying is lower than it has ever been. The barrier to not trying, however, has never been higher.

This is how failure becomes success.