AiCESoftware Engineering Foundations →

Software Engineering Foundations · African-First Product Design

Learning from Success

M-Pesa as a product design case study, alongside African product successes and failures.

671 words · about 4 min read

Read this book

Read or download the PDF

Learning from Success: What African Products Got Right AiCE Africanism & Pan-African Studies — M-Pesa as a design case study M-Pesa: The Design Principles Behind the Revolution An earlier unit examined M-Pesa's development and impact as an economic and technological story. Here it is examined specifically as a product design case study — asking what design decisions made it succeed where dozens of subsequent fintech products in Africa failed to reach comparable scale. First, M-Pesa was designed for the actual user, not an imagined user. The product team — which included Kenyan engineers and researchers who understood the Kenyan market from the inside — conducted extensive research into how Kenyans actually managed money, what problems they urgently needed solved, and what constraints they operated under. They discovered that the most urgent financial need was not investment or savings, as a Western financial designer might assume, but the safe, affordable transfer of money between family members in different locations — a mother in Nairobi sending money to a child at university in Kisumu, or a worker in the city sending money home to a rural village. Second, M-Pesa used existing infrastructure. It worked on any mobile phone — not just smartphones, but the basic feature phones that the majority of Kenyans already owned. It used the SIM card toolkit that all mobile phones already had. It required no new hardware, no new devices, and no new technical knowledge beyond what any mobile phone user already possessed. Third, M-Pesa solved the physical cash-in/cash-out problem through an agent network — local shopkeepers and kiosk owners who acted as human ATMs. This was a design insight of genius: instead of trying to build a network of expensive bank branches or ATMs, M-Pesa turned every corner shop into a bank branch. The agent network was self-financing — agents earned a small commission on every transaction — and self-growing, because being an M-Pesa agent was a profitable business for the shopkeeper. © Adesina Olagunju | African Institute for Computing Excellence | 1 Fourth, M-Pesa was priced for the actual market. Transaction fees were set at levels affordable for even low-income users, and competitive with the informal money transfer agents who were the existing alternative. The product did not require a minimum balance or a monthly fee. Users paid only when they used it. Case Studies: Successes and Failures Product Outcome & Design Lesson M-Pesa (Kenya, 2007) 51 million users across 7 countries. Designed for actual users on actual devices; used existing infrastructure; human agent network; affordable pricing. Twiga Foods (Kenya) Direct smallholder-to-retailer supply chain serving 15,000+ farmers. Eliminated a specific inefficiency (exploitative middlemen); mobile-first. Zipline drones (Rwanda) Hundreds of thousands of deliveries across 4 countries. Solved a specific, urgent problem where existing road infrastructure failed. Hello Tractor (Nigeria) Tractor rental marketplace serving thousands of farmers. 'Uber for tractors' on basic phones; solves affordability and underutilisation together. Facebook Free Basics (multiple countries) Used by millions but banned in several countries. Design success but policy failure — underestimated African governments' sensitivity to digital sovereignty. Jumo (Pan-African lending) Significant users but high non-repayment in some markets. Imported Western credit-scoring models that did not account for African income volatility. Various e-commerce platforms Multiple high-profile failures. Assumed Western logistics; African informal address systems need completely different fulfilment models. Various app-only products High abandonment in rural areas. Designed for smartphones and broadband; failed to account for feature phone and 2G/3G reality. © Adesina Olagunju | African Institute for Computing Excellence | 2 Key Takeaways M-Pesa succeeded through four specific design decisions: solving the actual problem users had (not the problem a designer assumed they had), using existing infrastructure rather than building new hardware dependency, solving the cash-in/cash-out problem through a self-financing agent network, and pricing for the actual market. The failures alongside it — Jumo's imported credit models, various e-commerce platforms' assumed logistics — show the same pattern in reverse. The next lesson turns from studying others' research to conducting your own. © Adesina Olagunju | African Institute for Computing Excellence | 3