When Hesitation Costs More Than Action

The AI advantage for SMEs is being built right now, not in two years

85% of German SMEs think AI is important. Only 15% are acting on it. That's a 70-percentage-point gap. And it's growing.

The AI advantage for SMEs is being handed out right now, and not evenly. Companies that make a structured start today are building a lead that will prove to be a strategic asset in three to five years. Those who wait will pay more for consulting in two years, and start from a weaker position.

This isn't a forecast. These are the numbers.

What German SMEs are doing right now

According to the KfW SME Panel and Bitkom 2026, only 20% of German SMEs use AI productively. On a scale of 0 to 10, they rate AI's role in their own business at 1.6. For the next five years, they expect 4.1.

That's the ambition of a student aiming for a passing grade.

At the same time, companies worldwide plan to double their AI spending in 2026, from 0.8% to 1.7% of revenue. Budgets are growing. Results aren't, yet.

42% of companies abandoned most of their AI projects in 2025. These aren't companies that didn't care about AI. These are companies that started wrong.

Why so many SMEs haven't started yet

The reasons for the hesitation are real, but not always rational.

Unclear ROI: AI investments rarely show up on the P&L in the first six months. That's not a sign it isn't working. It's the normal curve for any new technology investment. The question isn't "Is this paying off yet?" It's "Will I be better positioned in 18 months than I am today?"

Missing expertise: Only 20% of SMEs use AI productively. That creates a trap: nobody around you has done it yet, so it's unclear how to start.

Wrong expectations: Treat AI as a complex technology decision and you'll never get moving. Getting started with AI doesn't need a data science team or high-end IT infrastructure. It needs a concrete problem and the right setup to work on it.

The cost of waiting

McKinsey estimates the value-creation potential for German SMEs through AI at 30 billion euros. That potential won't land evenly. It goes to whoever builds a working process early enough.

Companies that act in 2025 and 2026 gain a three-to-five-year competitive edge. That happens because expertise compounds. A team that has built one AI prototype for its own problem understands more than a team that watched ten webinars, and can move faster and more independently on the next problem.

That expertise gap is hard to close. The competitor who starts today won't just have better tools in two years. They'll have a team that knows how to work with AI.

The most common mistake when starting

Many companies that start with AI start wrong. They buy licences, book training, and hope for adoption. Adoption without direction doesn't generate a return.

The Section AI Proficiency Report shows it clearly: only 15% of AI use cases in companies are likely to generate ROI. 85% don't. Not because the tools are bad, but because nobody clarified beforehand which concrete problem needs solving.

So where to start with AI adoption isn't a technology question. It's a process question. Which problem costs your team the most time? Which task repeats every week and can be structured? Where's the most obvious potential?

These questions don't need a management consultancy. They need a structured space and the right people in the room.

AI strategy in Germany: what to learn from the frontrunners

Companies already using AI productively share one approach: they started small, but structured. Not with a 40-page AI strategy. With a concrete problem, a clearly defined goal, and a team that worked on it themselves.

The knowledge stays in the company. So does the motivation, because the result was their own project, not something an external agency delivered.

That's exactly what happens in the AI Design Sprint: in five days, your team identifies a concrete, relevant use case, builds a working AI prototype, and comes away understanding how AI works for their specific day-to-day. No strategy document, no vendor lock-in.

I cover how to turn this foundation into a broader AI strategy in my article on AI strategy for SMEs.

What still helps right now

The window for a real AI competitive edge won't close overnight. But it is closing.

Whoever makes a structured start in 2026 is still part of the frontrunner group. Whoever waits until a competitor has three successful projects done will pay more for the same starting point, and start from a weaker position.

The question "Should we really start now?" is settled. The real question is: where do we start?

Your next step

If you want to know which use case is the right entry point for your business, sign up for a free Discovery Call. In 30 minutes, we'll figure out whether and how an AI Design Sprint makes sense for you.

Bitkom / KfW SME Panel 2025/2026: 20% of SMEs use AI productively, self-assessment 1.6/10
S&P Global 2025: 42% of companies abandoned AI projects
Section AI Proficiency Report 2026: 15% of use cases likely to generate ROI
McKinsey: €30 billion in value-creation potential for German SMEs through AI

Back to overview