I was selling dried aromatic herbs into the European market. Peppermint, spearmint, lemon balm, lemon, Verbena, chamomile… Yes, growing chamomile in Ghana. Varieties I’d chosen because European buyers wanted them, not because they were easiest to grow. The margins were extraordinary. The buyers were committed. We had contracts that specified price, volume, and quality.
And it nearly destroyed the operation.
Building for a deal
What happens; I could grow the herbs. The buyers wanted them.
So we had to build up the entire chain on our site. Land, equipment, permits, certifications. The costs were staggering. But the geometry of the deal (committed European buyers, high margins, long contracts) made it rational.
What I didn’t account for: that infrastructure only worked if the deal held.
The EU buyers committed volume. We built facilities for that volume.
If you just passed, and then we started realizing the realities of Agriculture and what it means to be bullish: I often see the loss of nature state that if you plant a seed and water, it must grow…
I kept the facility. I diversified to other crops. But the lesson was clear: infrastructure built around a single buyer is infrastructure built on borrowed time.
The other side of the same problem
Then I watched it happen differently.
In other parts of Ghana, I saw maize farmers (thousands of smallholders) producing grain at scale. No infrastructure to dry it. No standards. No buyers waiting for certified commodity. The grain would sit in fields until post harvest losses consumed 20 to 30% of production. The farmers’ margins were destroyed not by price but by having nowhere to put their crop.
The constraint on their side was identical to the constraint on my side. The infrastructure wasn’t there. But in their case, the buyer wasn’t waiting either. A chicken-and-egg that nobody would break.
I saw this in Nigeria, where cassava processors couldn’t scale because input supply was unreliable. In Côte d’Ivoire, where cocoa beans rotted because aggregation infrastructure only existed seasonally. In Senegal, where groundnut farmers sold cheap to whoever showed up with a truck because they had no way to dry or store the crop.
What the pattern teaches
The pattern was the same everywhere. When market linkage infrastructure doesn’t exist, both supply and demand get stranded.
Farmers can’t scale because they have nowhere to put what they grow. Processors can’t scale because supply is inconsistent. Buyers can’t commit because processing is unreliable. Nobody is wrong. The system is just broken.
I watched this for fifteen years. Long enough to see individual actors get squeezed, long enough to see brilliant initiatives fail, long enough to see productive capacity sit unused because the connective tissue didn’t exist.
My own aromatic herbs operation taught me something specific: the problem isn’t just that the infrastructure doesn’t exist. The problem is that infrastructure built for one relationship, calibrated to one deal, scaled to one buyer’s volume, is always fragile. The moment that relationship shifts, the economics collapse.
The solution isn’t to build more infrastructure. It’s to build infrastructure that works for multiple relationships simultaneously. Supply diversified across many farmers. Processing shared across multiple crops and buyers. Logistics corridors that serve the region rather than a single facility.
The architectural conclusion
What I learned wasn’t agricultural. It was infrastructural. And it was systemic.
The missing piece wasn’t better seeds or more training or access to credit. The missing piece was architecture: a system designed so that supply, processing, and export worked as one integrated unit, not as three separate actors hoping the others showed up.
The aromatic herbs taught me that building infrastructure for a deal is building infrastructure for failure. The fifteen years that followed taught me what to build instead.
You need a system where the fixed costs are justified by volume from multiple sources, where the processing is designed to handle multiple crops across multiple seasons, where the buyer relationships are diversified enough that no single cancellation collapses the economics.
That’s not just a different approach to infrastructure. It’s a different theory of what infrastructure is for.
