ekonom.cz: Transforming a Rigid Company? In the Age of AI, It Is Better to Build a New One

When I talk to Czech business owners today, whether in traditional manufacturing or services, a similar pattern keeps repeating itself: the company is successful, but losing speed. Growth has brought more complex management, layers of processes and gradually accumulated IT systems.

AI transformation therefore makes sense, but in practice it is slow and demanding. Processes, technology and the way people work all need to change, and the historical inertia of large organisations slows everything down significantly. The data confirms this: according to McKinsey research, only around 1% of companies consider their AI transformation truly mature.

The Market Will Not Give You a Time-Out for a Long Transformation

When companies think about how to respond to the rise of artificial intelligence, most start the same way: they try to change what they already have. They map processes, introduce new systems, redraw organisational structures and train employees.

However, such a transformation often takes years and the outcome is uncertain. Every company has its own history: processes that took years to develop, IT systems that accumulated layer by layer, and people accustomed to working in a specific way. Large-scale change then runs into natural inertia.

Although according to Zebracat statistics around 72% of companies actively use AI today, most do so only in a partial way. It is not a genuine transformation, just adoption into isolated processes and activities. It is therefore no surprise that macro productivity is currently growing by only around 2% annually.

A Faster Lane: Building a Digital Twin From Scratch

In several companies where I act as an investor, we have therefore started trying a different approach. Instead of a complex rebuild of the existing organisation, we built a new one alongside it. Not as a replacement, but as a faster lane. The existing company keeps doing what it does well: manufacturing, selling and serving customers. Next to it, however, a smaller organisation emerges that is designed from day one for a world where data, automation and artificial intelligence are a natural part of work.

In the global business world, this is referred to as a digital twin. In the new company, entirely new systems are built and thought about in a completely different way, as if the original company were being founded again from scratch. Of the same size, but with the knowledge that the owners and management have accumulated about running a business. The focus is on a new company strategy, the principles and processes of how it operates, the roles within it and the outputs for the customer. Sales, marketing, logistics and financial management typically change completely.

Sometimes existing software for specialised departments such as finance and product development is retained, but through AI it is connected and improved to serve the new strategy. The biggest problem with existing companies is that they are locked into old processes and old roles. The result is something like a live demo. Then the hardest part begins: working with people, including teaching them their new roles or replacing them where necessary.

The Advantage of a Company Without History

A new company has one enormous advantage: it carries no historical baggage. It has no technical debt and no processes that need to be painfully rewritten. From day one it can operate with AI as a natural part of work, not as an add-on.

In practice, this means automated workflows, real-time analytics and systems that communicate with each other naturally. Many tasks that people previously did manually now run automatically. As a result, the team can focus on things that genuinely create value.

At the same time, the economics of growth change. A traditional company grows linearly: more customers means more employees, more managers and more complex coordination. An AI-native organisation, by contrast, scales primarily through systems. When more customers arrive, the company does not open a new recruiting round. Instead, it expands its automations, processes and digital agents. According to McKinsey estimates, AI could deliver up to 4.4 trillion USD annually in global productivity.

We should not fear that mass unemployment lies ahead. Quite the opposite. Thanks to falling technology costs, far more small and medium-sized companies can emerge. Internal systems that previously cost millions, from CRM to logistics to project management, can today be built for a fraction of the price. The pressure on cost reduction is already being felt by IT and software development companies. Tools such as so-called vibe-coding are significantly accelerating development and changing the economics of software creation.

Business as Code: Rules Run the Work, Not People

Alongside zero historical baggage, there is another advantage of the new company: processes written primarily for AI, not for people who each interpret them differently. In most companies today, processes exist in documents, emails or people’s heads. However, everyone understands them differently, so the result depends on whoever happens to be doing the work at that moment.

The approach I call Business as Code changes this. Company rules, from pricing to approvals, are converted into a structure that both humans and AI agents can work with. A single current source of truth emerges, with no need for interpretation.

The rules are connected similarly to layers of code: strategic decisions define the framework, operations flow from them, and AI always works with the current context. As a result, routine activities run consistently. For example, preparing contracts or proposals is reduced from hours to minutes, and people focus mainly on oversight and judgement.

The goal is not to replace people, but to change the economics of work. The system scales while people concentrate on judgement, relationships and development. This is precisely one of the key principles on which new, AI-native companies are built.

A Powerful Combination: Experience and New Architecture

This is where I see a major opportunity for Czech companies. Many of them are built on deep industry knowledge developed over decades. They understand manufacturing, customers and the market, and this is something technology startups are often still catching up on. If this experience is combined with a new technological architecture, the result can be a very powerful combination. The result can be a very powerful combination. While the legacy enterprise brings domain knowledge and customer relationships, this new entity delivers speed, technology, and a completely different way of working.

The point is not to destroy what entrepreneurs have spent years building. The existing business can continue to operate and generate value. Alongside it, however, a new one can emerge: smaller, more flexible and designed for a world where the ability to experiment and scale quickly is a decisive competitive advantage.

Today, the concept of the “renaissance of work”, meaning the transformation of companies in the age of artificial intelligence, is beginning to resonate in the Czech business environment. Our project experience is fairly clear: once such an experiment starts to work, it very quickly reveals new possibilities. As a result, these new companies often grow in ways that a traditional organisation built on people and historical processes simply cannot match.