← Back to the homepage

Our approach

The technology is no longer the real problem.

Anyone can open ChatGPT today.

AI can do a remarkable amount today. What matters is which part of it actually helps your company.

The harder questions start after that:

Where is AI worth it for us at all?
Which processes should we change?
Is our data ready for it?
What can we automate reliably?
What does it give us economically?
What changes for our team?

That is exactly where we come in.

Because you do not simply install AI in a company. You integrate it into existing work.

Tech + people.

AI changes more than software. It changes processes, decisions, responsibilities and the work of people.

That is why we look at both sides from the start.

TECH

Data · Processes · AI systems · Automation · Infrastructure · Prototyping

We examine what is technically possible, which prerequisites are missing, and how that can become a robust solution.

PEOPLE

Skills · Roles · Responsibility · Collaboration · Change

We look at who works with the solution, what experience already exists, and what people need to help shape the change.

Not building the technology first and taking people along afterwards. Both belong together from the beginning.

After project work with several hundred companies, the Boston Consulting Group puts it like this: around ten percent of the value of an AI rollout comes from the algorithms, twenty percent from the technology, and seventy percent from rethinking the human side.

Source: Boston Consulting Group

Tech + people. Not tech instead of people.

AI transformation happens on three levels.

ÄUSSERER WANDELUNTERNEHMENMENSCH

THE WORLD OUTSIDE

Technology · Market · Clients · Competition · Regulation

What is changing outside the company?

THE COMPANY

Business model · Processes · Data · Infrastructure · Services

What has to change inside the company?

THE PEOPLE

Knowledge · Experience · Skills · Responsibility · Roles

What does the change mean for the people who work with it?

A good AI strategy brings all three levels together.

How we work.

Understand. Analyse. Test. Implement. Measure. What happens in each step, and what comes out of it, is on the homepage.

See the process on the homepage →

So that this can be answered at the end, we record the baseline during the analysis. Many AI projects do not fail because they achieved nothing. They fail because nobody measured where things stood beforehand.

How we think about AI.

Technology changes fast. Our principles should therefore not depend on any particular tool.

01

Rethink instead of install.

AI changes work, not just tools. A new process should not simply be the old process with an AI step inside it. We look at how work can be organised better with the new possibilities.

02

Foundations before automation.

Good automation needs good foundations. Data, processes, access rights, infrastructure and responsibilities have to hold. Sometimes the best first AI measure is therefore not an AI measure yet.

03

Value before AI activity.

An impressive demo is not a business case. We ask not only what AI can do, but what actually gets better: time, cost, quality, capacity, risk or new services.

04

Strengthen the actual work.

Not automating as much as possible. Automating the right thing. An AI collects data, and data is not the same as experience. Without experience there is no intuition, and that is exactly what makes a company distinctive. AI should take load off people where routine and information work block their real abilities, and it should not blindly replace what makes your work yours.

05

Understand quickly. Decide deliberately.

Not chasing every hype. But not waiting until others are far ahead either. We want to understand new possibilities early, test them sensibly, and only then decide what belongs in the company for good.

06

Knowledge and independence are part of it.

Anyone who understands how a system arrives at its answers, and where its limits are, works with it more safely. That includes the question of what a company makes itself dependent on: which data sits where, who could switch it off tomorrow, and what can be run in house.

Position paper to follow.

We do not sell you as much AI as possible.

We are on the same side as you.

So for us, good advice also means being able to say:

We would not automate that yet.
You do not need AI for that.
Fix your data foundation first.
Standard software is entirely enough here.

Or: this is exactly where we should act now.

Our goal is not that your company ends up with as much AI as possible. Our goal is that your company is in better shape with AI than without it.