Why Most Leaders Are Building AI Teams for Yesterday’s Problems

I spent last week visiting a deep-tech manufacturer in Freiburg that shapes glass like plastic. Recyclable. At room temperature. Revolutionary stuff.

But here’s what struck me most: while this company is literally reshaping materials at the molecular level, most leadership teams are still hiring AI talent like it’s 2019.

The 70% Problem Nobody Talks About

Every morning on my walk, I listen to the YPO AI Brief. Yesterday’s episode hit a nerve: “The AI Hire Everyone Is Getting Wrong.” The core insight? Most companies are optimizing for the wrong 30% of AI transformation.

The BCG 10-20-70 rule breaks it down clearly:

• 10% of AI value comes from algorithms
• 20% from data and technology infrastructure
• 70% from process change, behavior change, and organizational redesign

Yet I keep seeing the same pattern across DACH SMEs: boards approve budgets for data scientists while protecting legacy workflows. They hire for technical prestige instead of transformation capability. They delegate AI to IT while avoiding the leadership decisions that actually matter.

From Problem-Solver to System-Builder

Recently, LinkedIn News featured my thoughts on leadership bottlenecks. The core lesson? Being helpful can quietly turn you into the constraint.

This connects directly to AI adoption. When leaders solve every AI exception themselves, they create what I call “efficient dependency loops.” Teams learn to escalate rather than iterate. AI pilots run for 18 months without touching real workflows.

The shift isn’t technical—it’s behavioral. From answer-giver to question-asker. From problem-solver to capability-builder.

The Glassomer Test

Back to that Freiburg factory visit. What impressed me wasn’t just the innovation—it was how they embedded radical technology into industrial-grade operations. No theater. No pilots that never ship. Just sustainable manufacturing that works at scale.

That’s the standard. Not whether your AI team can build impressive demos, but whether they can redesign workflows that actually run your business.

The Real AI Hiring Framework

Instead of optimizing for the 30%, here’s what I’ve learned works:

The 70% Hiring Filter:

1. Change Leadership (40%) - Can they redesign processes while maintaining operations?
2. Behavioral Systems (20%) - Do they understand how humans actually adopt new tools?
3. Governance Design (10%) - Can they build guardrails that scale?
4. Technical Competence (30%) - The part everyone already focuses on

This isn’t about hiring fewer technical people. It’s about hiring technical people who understand that technology serves transformation, not the other way around.

Why This Matters for Transaction Readiness

For SME owners thinking about exits, this becomes critical. AI isn’t just operational efficiency—it’s transferability. Buyers want predictable, governed, scalable operations. Not brilliant founders solving AI exceptions manually.

The companies getting this right are building what I call “Fit-for-Transaction” AI: standardized processes that work without the original architect. Systems that create value, not dependency.

Next Week Action Items

1. Audit your AI hiring specs - How much weight goes to technical skills vs. change management experience?

2. Map your current AI bottlenecks - Where do escalations flow? Who’s solving exceptions manually?

3. Run the 70% diagnostic - For each AI initiative, what percentage of effort goes to technology vs. process redesign?

4. Test one micro-process - Pick something small and redesign it completely, not just automate the current version.

5. Schedule “AI leadership time” - Block 2 hours weekly to learn the tools your team is implementing.

The companies that treat AI transformation as a leadership discipline—not an IT project—will be the ones that actually capture the value.

What’s the biggest AI hiring mistake you’re seeing in your industry right now?