Every pitch deck about African payments shows the same map: arrows flowing across the continent, clean corridors connecting sender to receiver, mobile money logos neatly arranged in a grid. It looks simple. Connect the APIs, plug into the rails, money flows.
I run these corridors for a living — 15 of them, across West, Central, and East Africa, with 20+ IMTO partners. And I can tell you: the map is the easiest part. Everything that actually determines whether money arrives is invisible on that slide.
Here’s what the corridors actually look like from the inside.
Every Corridor Is a Different Business
People outside payments tend to think of “Africa” as one market. Even people inside payments sometimes talk about “our Africa operations” as if it’s a single thing to manage.
It’s not. Sierra Leone and Senegal are three hours apart by plane, but operationally they might as well be different planets. The mobile money operators are different. The regulatory frameworks are different. The settlement cycles are different. The FX dynamics are different. The agent networks behave differently. The failure modes are completely different.
A corridor through UEMOA (Senegal, Ivory Coast, Togo, Mali, Benin, Guinea-Bissau) benefits from a shared currency — the CFA franc — and a common central bank. That simplifies FX but doesn’t simplify anything else. Agent liquidity, mobile money API reliability, and regulatory reporting requirements still vary by country within the zone.
A corridor through Guinea runs on a different currency, a different regulatory regime, and mobile money infrastructure that behaves nothing like its neighbors.
Running 15 corridors doesn’t mean running one process 15 times. It means running 15 variations of a process that looks similar on paper but is entirely different in execution.
The Partner Relationship Is the Product
If you ask most fintech founders what their product is, they’ll point to their app, their API, or their platform. From an operations perspective, the real product is the partner relationship.
Your mobile money aggregator determines how fast payouts land. Your IMTO partner determines how settlements are structured and how disputes are resolved. Your agent network determines whether cash-out works in the last mile. If any of these relationships break down — slow settlements, disputed transactions, poor API uptime — the end customer’s experience breaks down too.
Managing 20+ IMTO partners means managing 20+ different settlement formats, reference structures, commission models, and escalation personalities. Some partners send settlement files daily. Some send them weekly. Some send them when they feel like it. Some use unique reference numbers that match cleanly to your system. Some use references that mutate between their transaction record and their settlement file.
This is not an API problem. This is a relationship, governance, and process problem. And it’s what makes or breaks payout operations.
Liquidity Is an Operations Problem, Not Just a Treasury Problem
Most companies treat liquidity as a treasury function: make sure there’s enough money in the right wallets, in the right currencies, at the right time. That’s true, but it misses the operational reality.
Liquidity in African payment corridors is volatile in ways that treasury models don’t always capture. A mobile money operator might run low on float in a specific region on a Friday afternoon. An agent network might have liquidity in one city but not another. A sudden spike in remittance volume — say, before Eid or Christmas — can drain payout wallets faster than they can be refilled.
The operations team is often the first to see liquidity stress, because they see the failed transactions. If a payout fails because the mobile money wallet is empty, that’s an ops problem before it’s a treasury problem. The feedback loop between operations and treasury needs to be fast — hours, not days.
We run wallet funding logs and real-time monitoring for exactly this reason. If a wallet balance drops below threshold, the operations team flags it before transactions start failing. Waiting for the failed transaction to trigger the alert means the customer already had a bad experience.
Technology Is a Fraction of the Job
I’ve seen companies raise millions to build payment infrastructure in Africa, spend most of it on engineering, and then struggle operationally because nobody planned for the human layer.
The technology gets money from point A to point B. The operations team handles everything that goes wrong between those two points — and something always goes wrong. A transaction gets stuck. A reference doesn’t match. A partner disputes a settlement amount. A payout lands in the wrong account. A mobile money API returns an ambiguous status that could mean success or failure.
Each of these requires a human decision. And the quality of that decision depends on training, process documentation, escalation clarity, and governance frameworks. A reconciliation officer who doesn’t know the difference between a tolerance exception and a genuine dispute will either approve things they shouldn’t or escalate things that don’t need escalating. Both cost money.
This is why I’ve invested heavily in SOPs, exception classification frameworks, and daily sign-off routines. Not because I love documentation — because without it, a 20-person team operating across 15 corridors will make inconsistent decisions, and inconsistent decisions at scale become financial losses.
The Hardest Part Isn’t Launching a Corridor — It’s Sustaining One
Getting a corridor live is a project. Keeping it running reliably is an operation. These require completely different skill sets and completely different attention.
A launch is exciting. You sign the partner agreement, integrate the API, run test transactions, and go live. There’s a team, a timeline, a Go/No-Go meeting.
Six months later, that corridor is just one of 15. The partner’s settlement file format has changed and nobody updated the reconciliation template. The mobile money operator upgraded their API and introduced a new error code your system doesn’t recognize. The agent who handles cash-out in a key district left and his replacement hasn’t been onboarded to the process.
Nobody writes a pitch deck about this. But this is what determines whether the corridor actually works.
What This Means for Anyone Building in This Space
If you’re a founder building payment infrastructure in Africa: invest in operations early. Not after your Series A, not after you’ve launched your fifth corridor. Now. Your first ops hire should be someone who has actually operated in these markets, not someone who’s managed operations in a market where the infrastructure works.
If you’re an investor evaluating payment companies in Africa: ask about reconciliation. Ask about exception resolution timelines. Ask about settlement disputes. If the founder can’t tell you how their team handles a failed payout at 11pm on a Friday in Conakry, the operation isn’t as strong as the deck suggests.
If you’re an operator working in this space: document everything. The process in your head is worth nothing if it’s only in your head. Write the SOP. Build the sign-off template. Create the exception register. When you leave — and you will eventually — the operation should survive without you.
The corridors on the map look clean. The work behind them isn’t. That’s the job.
Taha El Ghrib runs cross-border payment operations across 15+ African corridors. He writes about payment infrastructure, operational scale, and what actually breaks at TheOpsDesk.co.