AI Bootcamp in the Sweet Factory

There are moments in consulting projects when the setting becomes the perfect metaphor. The AI Bootcamp with SWEET TEC GmbH was one of those moments.

The most unusual setting for strategic AI work

An entire wall filled with hundreds of sweets. In every colour, shape and flavour. The parallel was impossible to miss: this wall looked exactly like the current AI landscape that leadership teams are confronted with right now.

Perfect-looking options everywhere Dozens of specialised tools for every conceivable use case

And how do most companies respond? They buy Co-Pilot licences for all employees. Before they even know exactly what for. After all, nobody wants to fall behind. The competition is doing it too. Every week a new, more powerful model comes onto the market. Claude Sonnet 4, GPT-5, Gemini Ultra 2.0.

The result? A bag full of all sorts of things. But no strategy for anything.

What matters is ignoring what glitters and focusing on what counts. The Sweet Tec leadership team faced many uncomfortable questions that must be answered before any tool decision:

1. State of play Where do we really stand – without sugar-coating and without “it’s the same for everyone else”?

2. Rules of the game How exactly does AI change OUR business? Not the confectionery industry in general. But Sweet Tec. With our products, margins, customers and very specific headaches.

3. Tool or transformer Is generative AI just a faster Excel for us – or a lever that could turn our entire business model upside down? 4. Our strategy What is our unfair advantage in the AI world?

Our launch pad was the question: Which three pilots will deliver savings or competitive advantages, i.e. added value, as early as tomorrow?

After an intensive day, the team had: ✓ A clear picture of where they really stand
✓ Their own strategy, one that fits their business
✓ Concrete pilots, ready to launch within weeks
✓ An honest look at the challenges that really matter

The question first, then the tool

The reflex is understandable. When a new model appears every week, waiting looks like falling behind. But a licence is not yet a strategy. Introducing tools before it is clear what they are for distributes possibilities, but no direction. The result is isolated experiments whose value nobody can assess.

The four questions the Sweet Tec leadership team faced reverse the order. First comes an honest assessment of where you stand, then an understanding of how AI changes your own business, and only then the question of tools. That way, the tool decision becomes a consequence of the strategy rather than a substitute for it.

Tool or transformer

The third question is the most uncomfortable. For many companies, generative AI is initially a faster tool: texts, analyses, evaluations. That is legitimate and often the right way in. It becomes dangerous when this view remains the only one.

Because AI can also call the business model itself into question: how customers buy, which services are still paid for, where competitors suddenly gain advantages. A leadership team should consciously think through both perspectives. Where does AI make us more efficient? And where could it fundamentally change our current position? Anyone who asks only the first question may end up optimising a model that is losing relevance.

Why three pilots

Limiting yourself to a few pilots is not a lack of ambition but a matter of focus. A good pilot is small enough to start within weeks and relevant enough to show real added value: a saving or a competitive advantage the team can name.

A common mistake is to choose pilots by technical feasibility rather than strategic importance. More helpful are questions such as: What specific problem does this solve? How will we know after a few weeks whether it works? And what do we learn from it for the next steps? The result is a short list of ready-to-launch initiatives rather than a declaration of intent.

What leadership teams can take away

• An AI strategy begins with an honest assessment of where you stand, not with buying licences.
• What matters is how AI changes your own business, not the industry in general.
• Leadership teams should think of AI both as an efficiency tool and as a potential attack on the business model.
• A few clearly justified pilots achieve more than many unconnected experiments.

Find out here how an AI Bootcamp takes your team from its own strategic question to ready-to-launch pilots. Explore the AI Bootcamp →

Source: Original on erwinhettich.com