A card is a solution to someone else's problem
Card networks were built for places with dense bank branches, near universal bank accounts, reliable electricity and merchants who could afford terminals. In much of Africa, none of those conditions held for most people. A great many adults had a phone before they had a bank account, and a phone, it turned out, was a better place to start.
That is the short answer to why mobile money still beats cards across much of the continent. The longer answer is about economics, trust and design, and it carries lessons for anyone building payments here.
Starting from the phone
Mobile money services let a person store value on their phone number, send it to another number and cash in and out through a network of agents. Kenya's M-Pesa, launched in 2007, is the best known example, and similar services followed in many countries across East, West and Central Africa.
The design was clever in what it did not require. It did not need a bank account, a branch visit, a smartphone or even a data connection, since the earliest services worked over simple menus on basic handsets. A person needed a phone, an identity document and a nearby agent.
Compare the steps a card asks for. A bank account first, with its documents and sometimes minimum balances. Then a card, then a place that accepts it, then a connection that works at the moment of payment. Each step drops out some of the people who might have been customers.
Agents are the real infrastructure
The visible product is an app or a menu. The real infrastructure is the agent network: small shops, kiosks and traders who convert cash to digital value and back. Agents put a financial service within walking distance of people who would otherwise travel hours to a bank branch.
Cash in and cash out matters because most economic life in these markets still runs on cash. A payment method that fits that world, and lets people move gradually from notes to digital balances, has an advantage that a card, which assumes you already live digitally, does not.
The economics for merchants
Cards charge merchants a fee on each sale, and accepting them needs a terminal or an online gateway, plus a bank account to receive funds. For a market trader or small shop with thin margins, that is a high barrier. Mobile money can often be accepted with nothing more than a number and a phone, and the transaction costs are easier to understand.
For customers, a transfer that arrives instantly and in the form they can use immediately, whether kept as a balance or cashed out, is simple and visible. It also works for the transactions that matter in daily life: sending money to family, paying school fees, buying airtime, settling small bills.
Trust did as much work as technology. A transfer confirmed by a text message, a receipt the agent can show, and a number to call when something fails: these small, visible assurances mattered to people who had good reason to be cautious about formal finance. Cards, with their statements and dispute processes, assume a customer who already trusts the institution.
There is a network effect as well. Once most of the people you pay and are paid by use the same service, every new user makes it more useful. A card cannot easily replicate that among customers who do not hold one.
Where cards still matter
None of this means cards are irrelevant, and the picture is not the same everywhere.
- Online and international payments still lean heavily on cards, because merchants abroad expect them.
- Higher income and urban customers often hold cards and use them readily.
- Nigeria is a different case. Mobile money wallets exist, but instant bank transfers, using account numbers and bank apps, handle a large share of everyday digital payments. The gap that mobile money filled elsewhere was met here by banks and transfer rails.
- Regulation shapes outcomes. Who may issue wallets, how agents are licensed and how rails connect to each other differ by country.
So the claim holds best where bank accounts were scarce and mobile operators led, and less well where banks moved fast.
What builders should take from it
First, design for the customer's real starting point. If they do not have a bank account, do not make one a prerequisite. If connectivity is patchy, make the flow work with interruptions.
Second, interoperability decides who wins in the long run. Services that let money move between wallets, banks and cards grow faster than closed ones, because users do not want to hold five balances.
Third, trust is built through reliability and recourse. People will use a system with a clear way to resolve a failed transaction, and they will leave one that loses money without explanation.
Fourth, support the whole journey, including cash. Pretending the economy is already digital is the quickest route to building something nobody uses.
Not a contest
It is tempting to frame this as a contest in which one method defeats another. The better reading is that each serves a different set of conditions, and the winner in any market is the option that meets people where they are. In much of Africa, that has been the phone and the agent down the road.