The verdict in three sentences
Reconciling by hand means comparing a bank export spreadsheet to your orders line by line: 4 to 8 hours lost each week and gaps that go unnoticed. The right decision in 2026 is automatic matching by unique reference between each mobile money transaction and its order, with a tolerated variance of 0 FCFA. Anything unmatched within 24 h triggers an alert, and a daily report gives you the match rate.
The reference-matching principle
For every order you generate a unique reference (e.g. KOL-2026-04517) passed to the payment provider. The settlement (Wave, Orange Money, Paystack) returns that same reference. The reconciliation engine matches transaction to order on this key, checks the expected net amount, and classifies the rest as anomalies.
| Status | Condition | Automatic action |
|---|---|---|
| Matched | Ref + amount OK | Order marked paid |
| Amount variance | Ref OK, net ≠ expected | Alert, check fees |
| Provider orphan | Transaction with no order | Investigate within 24 h |
| Order orphan | Order with no transaction | Chase or cancel |
| Duplicate | 2 transactions, 1 ref | Refund the duplicate |
2026 data and KPIs
In Nigeria, Paystack settlement arrives at T+1 and fees explain most amount variances. The right metric is not "zero anomaly" but a match rate ≥ 99 % with orphans handled within 24 h.
| Metric | 2026 target | Frequency |
|---|---|---|
| Match rate | ≥ 99 % | Daily |
| Tolerated amount variance | 0 FCFA (excl. known fees) | Per transaction |
| Paystack settlement delay | T+1 | Per batch |
| Paystack local fee | ~1.5 % | Per transaction |
| Unmatched alert | > 24 h | Automatic |
| Manual time before | 4-8 h/week | — |
| Manual time after | < 30 min/week | — |
| Daily report | 1 email / dashboard | 07:00 |
The daily report lists: transaction count, gross amount, total fees, net credited, orphan count and age of the oldest unmatched item.
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
Mini case study
Ibrahim, a restaurateur in Abidjan, handles 1,200 payments/month (Wave + Orange Money). Manually he spent 6 h/week matching, i.e. ~24 h/month; at 2,500 FCFA/h of his time, that is 60,000 FCFA/month "burned", and he found 3 to 5 uncredited transactions each month. With automatic matching, his time drops to 2 h/month and orphans are cleared same-day: he recovers on average 45,000 FCFA/month of previously lost collections.
FAQ
Do I need a unique reference per order? Yes, it is the pivot of the whole reconciliation. Without a unique key passed to the provider, matching falls back to amount + timestamp, far less reliable as soon as baskets are identical.
How do fees fit into the amount variance? Expected net = gross − known fees (e.g. ~1.5 %). You tolerate 0 FCFA variance on the computed net; any divergence flags an unexpected fee or an error.
What about a provider orphan? That is a transaction with no order on the site: often a lost webhook. You match it within 24 h and create/validate the corresponding order before it becomes a dispute.
Does T+1 create a cash-flow problem? Paystack T+1 settlement means the money is credited the next day; reconciliation runs on the previous day's batch, which does not prevent confirming the order in real time via the webhook.
Can several providers be reconciled together? Yes: a single engine ingests Wave, Orange Money and Paystack exports, normalizes the columns and produces a consolidated match rate.
Let's talk about your project. We set up your multi-provider reconciliation engine with a daily report and sub-24 h alerts. 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.
