What Happens When the Founder’s Eye Is the Only Quality Control — A Case Study in Gorgeous Dysfunction
Erwee Coetzee
Diamond Stack
Cape Town, South Africa
Thandi built a reputation on exquisite taste.
For 12 years, her Cape Town atelier produced pieces that sold themselves. Collectors sought her out. Galleries competed to stock her work. Every stone she selected was perfect. Every design was distinctive. Every piece commanded a premium.
By year 12, she had five artisans, two apprentices, and a waiting list that stretched six months.
The obvious next step: delegate the sourcing.
One of her artisans—let’s call him Mthembu—had worked with her for seven years. He knew her taste. Or so she thought. “Mthembu,” she said, “I’m going to trust you with the stone selection for the next collection. I’ve taught you everything I know. You can do this.”
Mthembu was thrilled. Finally, recognition. Finally, real responsibility.
Thandi spent the next month on strategy: new markets, wholesale partnerships, brand positioning. She checked in on Mthembu’s progress once a week. “How’s the sourcing going?” “Great,” he’d say. “Really great. You’ll love what I found.”
When the pieces arrived four weeks later, Thandi’s heart sank.
The stones were fine. Technically fine. Good colour, good clarity. But they were boring. Safe. The kind of stones that any jeweller would choose. Not the kind Thandi would choose. Not the kind that commanded premiums. Not the kind that made collectors’ eyes light up when they saw them.
She pulled Mthembu aside. “These aren’t right. I need you to understand—the difference between a good stone and the right stone isn’t something I can teach. It’s intuition. It’s an eye you either have or you don’t.”
Mthembu’s face fell. He’d spent four weeks sourcing. He’d done everything she’d asked. And it still wasn’t good enough.
So Thandi made a decision. She’d do the sourcing herself. Just for now. Until she figured out how to teach the intuition.
By year 14, she was working 14-hour days again. Sourcing, designing, approving everything. Her “delegation experiment” was over.
By year 15, her best artisan left to start his own atelier.
By year 16, her reputation started to slip. The waiting list shrank. Galleries asked for discounts. Collectors moved to other makers.
By year 18, Thandi’s atelier closed. She couldn’t afford to keep it open.
The Missed Intervention Point
This is a true story. The details are anonymised, but I watched this happen. And what breaks my heart is this: Thandi’s failure was predictable. It was also preventable.
The problem wasn’t that Mthembu couldn’t learn. The problem was that Thandi was trying to delegate the wrong thing.
George Akerlof won a Nobel Prize for describing a concept called information asymmetry (Akerlof, 1970). In simple terms: when a buyer can’t easily verify the quality of what they’re buying, the market breaks down.
In Thandi’s case, she had knowledge that Mthembu didn’t have. Not knowledge of facts—not “these stones are VS clarity” or “this cut is princess.” That’s explicit knowledge. Easy to teach.
What Thandi had was tacit knowledge. The gut sense of which stones have presence. The intuition about which designs will resonate. The aesthetic judgment that comes from 12 years of observation.
Nonaka and Takeuchi (1995) spent their careers studying tacit knowledge—the stuff that lives in your hands and your intuition, not in your textbooks. Their conclusion: tacit knowledge can’t be delegated. It has to be externalised.
There’s a difference. Delegation says: “You do this now.” Externalisation says: “Let me show you how I think about this, so you can develop your own judgment.”
Thandi tried to delegate stone selection. What she should have done is externalised her decision-making process.
She should have said: “Mthembu, we’re going to source stones together for the next three months. Every time I pick a stone, I’m going to tell you why. Not the technical specs—the reasoning. Why this one has presence and that one doesn’t. Why this colour works for the market and that colour sits in a weird middle zone. Let’s build your intuition together.”
That’s externalisation. That’s how tacit knowledge gets distributed.
But it requires something delegation doesn’t: the founder’s active participation in the learning process.
Thandi wanted to step back. Understandable—she was exhausted. But stepping back too fast killed the knowledge transfer. Mthembu didn’t fail because he was incapable. He failed because the method of knowledge transfer was broken.
Why This Matters for Your Business
Post 1 gave you credibility through honesty. Post 2 gave you permission—it’s not your fault you’re a bottleneck. Post 3 (this one) is the cautionary tale: but ignoring the bottleneck will destroy your business.
Thandi’s story isn’t unique. I’ve seen it play out in a dozen different ateliers, in different cities, with different founders. The pattern is always the same:
Year 1–8: Founder does everything. Quality is exceptional. Business grows.
Year 9–10: Founder realizes they’re a bottleneck. Tries to delegate.
Year 11–12: Delegation fails because tacit knowledge can’t be transferred that way. Founder takes decisions back.
Year 13–15: Founder is exhausted. Team is demoralized. Best people leave. Quality slips.
Year 16+: Business either dies or shrinks back to founder-only model.
This doesn’t have to happen to you. But it will happen if you ignore the structural reality of the bottleneck.
The Intervention That Works
Here’s what Thandi should have done—what you can do—to break the cycle:
Step 1: Acknowledge the bottleneck is structural. Not a sign of weakness. A sign that your business has grown past a single founder.
Step 2: Don’t delegate decisions. Externalise your reasoning. Bring your team into the thinking process. Show them how you decide, not just what you decide.
Step 3: Build decision frameworks together. Over time (three months, six months, a year), turn your tacit knowledge into explicit frameworks. “Here’s what I look for in a stone.” “Here’s how I think about pricing.” “Here’s what quality means to me.”
Step 4: Let them practice within the framework. Now that they understand your reasoning, they can make decisions—not perfectly like you, but well enough. And they’ll get better with feedback.
This takes longer than simple delegation. But it works. Because you’re not trying to clone your intuition. You’re trying to build judgment in your team.
What Post 4 Will Show You
Post 3 is the “what goes wrong” post. It’s the cautionary tale. It’s meant to make you sit up and say: “Oh. That could be me. That is me, actually.”
Post 4 will be the “how it could have been solved” post. It will show you the exact decision Thandi should have made at year 9. The moment where the atelier could have been saved.
Then Posts 5–8 will show you the systematic way to build that decision framework for every area of your business—not just stone selection, but pricing, design direction, customer conversations, team management.
Your Reflection for This Week
Last week I asked: What decisions are you making that could be made by your team if they had the right framework?
This week, I want you to sit with something harder:
Where are you delegating decisions when you should be externalising your reasoning? Where are you expecting your team to just know what you know?
Stone selection. Design direction. Pricing. Customer conversations. Employee discipline. Which of these are you trying to push down the hierarchy without actually teaching how you think about them?
That’s where Thandi’s business cracked. That’s where most ateliers crack. And that’s exactly where we’re going to build in Posts 5–8.
Research & References
Akerlof, G. A. (1970). The market for “lemons”: Quality uncertainty and the market mechanism. The Quarterly Journal of Economics, 84(3), 488–500.
Foundational work on information asymmetry. Explains why markets break down when buyers can’t verify quality. Directly applicable to the founder’s tacit knowledge problem: the founder has quality judgment that the team can’t verify.
Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating company: How Japanese companies create the dynamics of innovation. Oxford University Press.
The definitive work on tacit vs. explicit knowledge. Provides the theoretical framework for understanding why founder intuition can’t be delegated—it must be externalised through dialogue and practice.
Feldman, M. S., & Pentland, B. T. (2003). Reconceptualizing organizational routines as a source of flexibility and change. Administrative Science Quarterly, 48(1), 94–118.
On the difference between ostensive routines (the prescribed way) and performative routines (the enacted way). Relevant to externalisation: as team members practice within the founder’s framework, they adapt it to their own context.
