The verdict in three sentences
Mobile money fraud is not a wall, it's a leak: for a Nigerian e-commerce store, it represents in 2026 an order of magnitude of 0.3% to 1.2% of volume depending on average basket and sector. The classic mistake is locking everything down, which scares away more real customers than fraudsters. The right approach combines velocity rules, a risk score and targeted manual review on fewer than 3% of orders.
The signals that reveal a risky transaction
Mobile money fraud in West Africa follows recognisable patterns. A burner number created minutes ago, an account name that doesn't match the delivery name, five payment attempts in two minutes: each signal alone proves nothing. Combined, they push the score up.
The principle: each signal adds risk points. Below a threshold, the order passes. Above it, manual review. Well above, automatic decline.
| Detected signal | Risk points | Typical false positives |
|---|---|---|
| More than 4 payments in 2 min (velocity) | +30 | Customer retrying after timeout |
| Mobile money number created < 24h | +25 | Legitimate new customer |
| Account name / delivery name mismatch | +20 | Buying for a relative |
| IP outside the operator's country | +15 | Diaspora, VPN |
| Basket > 5x average basket | +15 | Genuine group purchase |
| 3rd refund requested in 30 days | +35 | Legitimate logistics dispute |
| Disposable email (temp domain) | +20 | Rare among real customers |
Decision thresholds and the cost of a chargeback
Once the score is computed, you need clear thresholds. Too low and the team drowns in reviews; too high and fraud slips through. Here is a 2026 reference grid for a mid-sized merchant.
| Total score | Decision | Share of orders | Average cost per case |
|---|---|---|---|
| 0 - 34 | Auto-accepted | ~90% | 0 FCFA |
| 35 - 59 | Manual review (< 10 min) | ~7% | 1,500 FCFA (agent time) |
| 60 - 79 | Step-up verification (OTP, call) | ~2% | 3,000 FCFA |
| 80 and above | Auto-declined | ~1% | 0 FCFA (avoided) |
| Confirmed chargeback | Hard loss + fees | 0.3 - 1.2% | 8,000 - 45,000 FCFA |
A chargeback costs more than the basket: the product already shipped, dispute-handling fees (often 5,000 to 10,000 FCFA per case) and the risk of penalties if your rate exceeds the aggregator's threshold.
Mini case study
Ada runs a cosmetics store in Lagos, 4,000,000 FCFA (~2.5m NGN equivalent) in monthly volume. Before antifraud rules she suffered 0.9% fraud, or 36,000 FCFA/month in losses plus 3 chargebacks at 8,000 FCFA in fees = 60,000 FCFA/month. By turning on velocity checks and a risk score she blocks 80% of cases, cutting the loss to ~12,000 FCFA/month. Manual review costs her 2h/week (~24,000 FCFA/month), but the net gain stays positive at ~24,000 FCFA/month, not counting reputation preserved.
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
FAQ
What fraud rate is "normal" for mobile money in 2026?
Between 0.3% and 1.2% of volume depending on sector: cosmetics and electronics are more exposed than groceries. Above 1.5%, rules need urgent review.
Will the risk score block my genuine customers?
Properly calibrated, no: the target is keeping auto-declines under 1% of orders and manual review under 3%. False positives are fixed by tuning points per signal.
What does a chargeback really cost?
Between 8,000 and 45,000 FCFA depending on basket, including the lost product and case fees (5,000 to 10,000 FCFA). A high rate can also trigger aggregator penalties.
Do I need an OTP on every payment?
No, only on mid-range scores (60-79). A blanket OTP adds friction and drops conversion by 3 to 8% without cutting fraud much.
Can I automate all this without a data scientist?
Yes: a weighted rules grid covers 90% of cases. Machine learning only pays off above a certain volume (order of magnitude: 15,000 orders/month).
Let's talk about your project. We integrate antifraud scoring tuned to your mobile money volume, without breaking conversion. WhatsApp +221 77 596 93 33.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.
