I've built software in capital markets, in healthcare, in a two-sided marketplace, in print and manufacturing, and in retail automation. Different industries, different customers, different problems. And after enough cycles of building, launching, and learning the hard way, I've noticed something: the lessons that actually matter aren't the ones in the startup books. The books are written for San Francisco. I build in Harare.
So here's what I wish someone had told me — the things nobody warns you about when you build software for African businesses. None of this is theory. All of it cost me something to learn.
1. Your beautiful app is competing with WhatsApp, and WhatsApp is winning
The first hard lesson: your customer doesn't want your app. They want their problem solved in the place they already are — and that place is WhatsApp.
I've watched founders (myself included, early on) pour months into a polished mobile app, only to find customers ignore it and message a WhatsApp number instead. Not because the app was bad, but because downloading, registering, and learning a new interface is friction, and WhatsApp is zero friction. People already live there.
The lesson isn't "don't build apps." It's "meet people where they are before you ask them to come where you are." The best African software often looks less impressive than a Silicon Valley product and works far better for its market, because it rides rails people already trust — WhatsApp, USSD, mobile money — instead of demanding they adopt new ones. Ego wants the slick app. The business wants the boring thing customers actually use.
2. "It works on my machine" means nothing when the machine is a $60 phone on 3G
In the demo, everything is fast. You're on fibre, on a flagship phone, in a well-lit room. Your customer is on a $60 Android, on patchy 3G, with 2% battery and a data bundle they're rationing.
If your product assumes reliable connectivity, generous data, and modern hardware, it will fail silently for most of the people you're trying to serve — and you won't even see it fail, because your own experience is fine. I learned to design for the worst realistic conditions, not the best: lightweight pages, offline tolerance where possible, flows that survive a dropped connection, interfaces that don't punish a slow phone.
Building for constraint isn't a downgrade. It's the actual skill. Anyone can build something that works on fast infrastructure. Building something that works on African infrastructure is the harder, rarer, more valuable craft.
3. Trust is the product. The software is just how you deliver it.
This is the one that took me longest to internalise. In markets that have lived through bank failures, currency collapses, and more than a few scams, people don't extend trust cheaply — and no feature list will earn it for you.
When you're handling someone's money, their health records, or a transaction between two strangers, the technology is the easy part. The hard part is convincing a human being that you are safe, real, and still here tomorrow. That means visible signals of legitimacy, responsiveness when something goes wrong, and consistency over a long time. I've seen technically inferior products win because people trusted them, and technically superior products die because people didn't.
If you're building for African businesses, you are not in the software business. You're in the trust business, and software is your delivery mechanism. Act accordingly — over-communicate, honour every commitment, and treat a customer's confidence as the actual asset you're accumulating.
4. The "informal" economy is not disorganised. It's differently organised.
Outsiders look at African markets and see chaos — cash, informal credit, handshake deals, no paperwork. So they build software to "formalise" it, and it flops, because they mistook unfamiliar for broken.
The informal economy runs on real systems: reputation, community, trust networks, oral records that everyone honours. A market trader who keeps no written books may know her margins, her regulars, and her supplier terms with total precision. Software that ignores or overrides those systems feels hostile. Software that digitises and strengthens them feels like a gift.
The lesson is humility. Before you "fix" how people do business, understand why they do it that way — because they've usually optimised for constraints you haven't noticed yet. The best products I've built worked with the existing grain, not against it.
5. You will be the founder, the engineer, the salesperson, and the support line
The startup mythology has specialists — a growth team, a design team, a support team. Building here, especially early, you are all of them, often on the same day. I've written production code in the morning, pitched a client at lunch, and personally walked a confused customer through a flow at night.
This is exhausting, and it's also the secret advantage. When you're the one who codes and the one who hears the customer's complaint directly, the feedback loop is brutally short. You feel the problem and you can fix it the same week. Big companies pay fortunes to get closer to their customers. You're already there. The trick is to stop resenting the many-hats reality and start using it — the direct line between building and selling is a moat, not a burden.
The thread that ties it together
If there's one thing underneath all five lessons, it's this: building software for African businesses is less about technology and more about deeply understanding a context that most of the tools, tutorials, and templates were never designed for.
The frameworks are global. The problems are local. And the gap between them is exactly where the opportunity lives — because the people who can hold both, who can write the code and understand why a market trader in Mbare does business the way she does, are rare. That combination is the real skill. Not the stack. Not the framework. The judgement to build the right thing for a market that the rest of the industry doesn't bother to understand.
That's the work. It's harder than the books make it sound, and far more worth doing.
