A trader in Lagos collects payment through OPay, pays a supplier through PalmPay, and keeps a separate app for the accounting tool she actually needs. Three financial apps, one small business, and no single place where the useful bits meet. Nigeria has become very good at moving money on a phone; it has not yet become good at letting the long tail of services live where the money already moves.

The context is a country that other markets send delegations to study. DataReportal counted 109 million internet users in Nigeria at the end of 2025, a penetration rate of 45.5 percent, against 165 million active cellular connections across a population of 239 million. The median age was just 18.1 — the youngest of the major markets in this series. Agency banking and mobile wallets reached towns the formal banking system never served. The next layer, the one where a service other than payments gets used, is the part still waiting for a cheap distribution surface.

The rails are world-class

Nigeria's fintech reputation is earned. Moniepoint moves serious volume through an agent network, OPay and PalmPay put wallets in millions of pockets, and Kuda made the neobank feel ordinary. For a large merchant, the tooling is adequate. The problem sits below that tier: the hairdresser, the pharmacy, the insurance agent, the community group — economically active people who cannot justify a native app, and whom a native app cannot afford to acquire.

Player or signalWhat it does todayWhere a mini-program fits
OPayWallet and payments at scaleA service surface inside a trusted shell
PalmPayConsumer walletCheckout that stays in the flow
MoniepointAgents and SME bankingIn-app tools for small businesses
KudaNeobankEmbedded products, no new install
109M internet users45.5% of a 239M populationReach that starts before a download

The last row is the quiet constraint. With just under half the population online and social media identities at only 20 percent of the total, mass distribution through discovery feeds is not the default it is in more connected markets. The wallet app, meanwhile, is opened for reasons that have nothing to do with browsing. That makes it a far more valuable surface than its feature set suggests — a place where a service can be encountered at the exact moment a transaction is already underway.

Trust is earned in the shell

Here is the nuance that outside operators routinely get wrong. Nigerians have been burned by fintech apps that stalled, vanished, or quietly changed the rules. Trust in this market is not won by a new logo and a launch campaign; it is inherited from a shell that has already survived scrutiny. A mini-program sandboxed inside a wallet people rely on starts with a credibility its standalone rival must spend years building.

There is a harder edge to that argument, and it is worth naming. Containment protects the host as much as the guest. If a service misbehaves, the runtime can cut it off without the damage spreading into the wallet's own balance and transfer functions. That is precisely why a host says yes: the upside of a richer catalog comes with a clean, single-action exit. Any catalog design that ignores the exit will struggle to get admitted in the first place.

The long tail is the catalog

The instinct after seeing OPay's numbers is to build another wallet. That is the most crowded road in Nigerian fintech, and the one with the weakest reason to exist. The more interesting inventory is everything that does not need to be a wallet at all — booking, membership, micro-insurance, remittance top-ups, merchant tooling — delivered as services inside a shell that already handles identity and money.

A shared runtime is what makes that inventory economic. Build the payment adapter once, reuse the identity and consent path, and publish an update through one channel that reaches every host. The alternative is a dozen small teams each solving the same integration against a different app, and a market where the best service is the one that could afford the most installs. Nigeria has already proven it can build rails. The open question is whether the service layer arrives as a catalog or as another wave of apps that nobody finishes installing.

Sources