Digitalization projects in African markets fail for reasons that have very little to do with technology and almost everything to do with imported assumptions.
Start with infrastructure. Power is not continuous, and connectivity is neither continuous nor cheap. A system that syncs constantly is not merely inconvenient — it is expensive to the person using it, who is paying per megabyte. Designing for the network being absent, and for data itself being a cost the user bears, changes what a reasonable architecture looks like.
Then payments. A great deal of commerce settles in cash or through mobile money, not cards. Mobile money is not a worse card — it is a different instrument, with its own confirmation flow, its own failure modes and its own reconciliation lag. Systems built around card semantics model it badly, and the mismatch surfaces as accounting that never quite balances.
Device reality matters as much. The primary computer is a mid-range Android phone, often shared, frequently on a small screen with limited storage. An interface designed on a desktop and 'made responsive' is not the same as one designed for that device. Neither is a bundle that assumes an unmetered connection to download.
Underneath all of it is trust. Adoption depends on whether people believe the system will not lose their money, their records or their standing — and that belief is earned through visible confirmation, human recourse when something goes wrong, and the ability to see and hold a record. A digitalization effort that ignores this ships a technically correct product that nobody uses.
The conclusion is not that African markets need simpler software. It is that they expose assumptions the rest of the industry gets away with. Systems built to work here tend to be more robust everywhere.