"What are your fees?" is every merchant's first question, and it is the right one. But a rate on its own answers about a fifth of it. The cost of collecting money has five lines, and only one of them appears on a pricing page. Here are all five, with what you need to work out your own.
Line 1 - The collection commission
This is the one everybody looks at. At WalleoPay it is 3% of the amount, taken only on payments that succeed. No subscription, no setup fee, no minimum billing amount, no fixed part per transaction.
On a sale of 10,000 FCFA, the commission is 300 francs and your balance is credited with 9,700. On 1,250 francs the arithmetic gives 37.5: since the CFA franc has no cents, rounding goes up, to 38 francs. It is written once, when the payment is created, and stored with it - the commission is never recomputed afterwards.
What matters for your margin is not the rate but its base: a declined, cancelled, expired or abandoned payment carries no commission at all. You pay on what you collect, not on what you attempt. That is a real difference from models where every attempt is billed.
Line 2 - Getting the money out
Your balance is credited in francs, and you request a payout to your Mobile Money number whenever you want, from 1,000 FCFA upwards. Nothing surprising so far.
Two costs come after that, and they are worth telling apart. The payout sending fee is shown to you before you confirm the request, so you know what will land before you click. Then the cash withdrawal at an agent, charged by your operator according to its own schedule. That schedule is public, displayed at agent counters and readable from your operator's menu. It works in bands, which has two very concrete consequences:
- Withdrawing once costs less than withdrawing five times. Five withdrawals of 20,000 francs almost always cost more than a single one of 100,000.
- Withdrawing is not always necessary. If you pay your suppliers, your riders or your bills in Mobile Money, money that stays in the wallet costs nothing. The cost comes from converting to cash, not from collecting.
Plenty of merchants discover, once they do the arithmetic, that their heaviest cost line is not the gateway commission but their habit of emptying the wallet in small notes every other day.
Line 3 - Reconciliation time
This is the invisible line, and often the most expensive one.
Collecting without automation - USSD merchant code, confirmation SMS, screenshot sent by the customer - costs no commission. In exchange, for every sale somebody has to read the SMS, find the order, compare the amount, check the time and mark the order paid. Count a minute per transaction when all goes well, five when you have to call the customer.
Run the numbers on your own volume. At 40 sales a month, that is under an hour: negligible. At 40 sales a day, it is an hour every day, around twenty hours a month - a part-time job nobody budgeted for. Value that hour at the real cost of the person spending it, and compare it with 3% over the same period. The answer often surprises.
On top of the time sits the cost of reconciliation errors: the order marked paid twice, the one delivered against a fabricated screenshot, the one forgotten because the SMS arrived during a power cut. Rare one by one, expensive in aggregate.
Line 4 - The sales that never complete
A failed payment costs you no commission. It costs you the sale.
It is the most profitable line to work on, because you do not have to negotiate it with anyone: it depends entirely on your payment journey. A customer whose wallet is empty, who never sees the confirmation prompt, or who does not understand the error message, is a lost sale at zero cost to the gateway and full cost to you.
Two concrete things make the difference: a failure message that says what to do - "top up at an agent and try again" rather than "payment declined" - and a button that resends the request without recreating the order. On most shops, recovering one failing customer in ten is worth more than three tenths of a point of commission.
Line 5 - Cash sitting still
Between the moment the customer is debited and the moment the money is available to you, time passes. That delay costs you something when it forces you to front your stock, and it is worth something when it is short.
So the question to ask any solution is precise: when is my money available, and who decides to release it? With a balance credited the moment a payment succeeds and paid out on request, the delay is yours to control. With a weekly settlement, it belongs to a calendar.
Six questions to ask before you sign
Whatever solution you are comparing, the answers to these six say far more about real cost than a rate printed in large type:
- Is there a monthly subscription, a setup fee, a minimum amount billed?
- Is the commission taken on attempts or only on successful payments?
- Is there a fixed part per transaction? On small amounts it weighs more than the percentage.
- Is a payout to my Mobile Money number charged, and from what amount can I request one?
- When does the money become available, and does that schedule depend on me?
- Is every movement traced line by line, with its commission and the resulting balance?
That last question sounds like a detail. It is not: a solution that will not show you the detail of what it deducts is a solution whose own pricing you cannot verify.
An order of magnitude
Take a shop doing 600,000 FCFA of sales in a month, across 40 orders of 15,000 francs. The collection commission comes to 18,000 francs, that is 3%.
Against that, manual collection on a merchant code shows 0 francs of commission - but implies 40 reconciliations, a dispute or two to settle, and as many cash withdrawals as trips to the agent. The tipping point is not a rate, it is a volume: as long as reconciliation fits into a few minutes a day, manual remains defensible; beyond that, it costs more than the commission it saves.
So the right instinct is not to hunt for the lowest rate, but to work out once, on your own figures, what each of the five lines costs you. Most merchants who do the exercise find the three middle lines outweigh the first.