Lenders
Underwrite stock you can see
You set the credit box, the rate table and the floor on the score. We bring the application, the history behind it and the paperwork that closes it.
For lendersSupply chain finance infrastructure
Stock today, paid for over the weeks it sells.
A shopkeeper asks for inventory over WhatsApp. A microfinance institution prices it against what that shop has actually bought before. The distributor is paid directly and releases the goods against a collection code. Fin&CGX is the rails and the score underneath.
One loan, end to end
Seven steps, and the shopkeeper only sees the first and the last. Everything between them happens where the money and the goods actually move.
The retailer asks
Over WhatsApp, in French or English. No app to install and no account to create — a phone number and a four-digit PIN is the whole of it.
The score answers
Built from invoices the distributor has already raised against that shop and that a person has verified. Trading history, not a credit bureau that has never heard of them.
A lender reviews and prices
A lending officer sees the request, the score and the shop’s file, and sets the amount and the schedule inside the credit policy their institution wrote.
The offer goes back
The cost, the instalments and the dates, in the retailer’s own language. They accept or decline on the same thread they asked on.
The agreement is co-signed
A contract naming the lender, the borrower and the distributor, issued as a PDF and signed by both sides before any money moves.
The distributor is paid, the goods are collected
The lender pays the distributor directly — no cash reaches the shopkeeper. They collect the stock against a code their counter checks.
The retailer repays
On the schedule they agreed, with reminders before each date, until the loan closes.
Three sides, one ledger
Distributors
Sell more, and get paid on release
Your customers buy the stock they could not fund this week. You are paid by the lender when the goods go out, not when the shop has sold them.
For distributorsRetailers
Nothing to install
Ask on WhatsApp. Collect at the counter you already buy from. Repay on a schedule you agreed before you took the goods.
For retailersA score that admits what it does not know
Most of the shops here have no file anywhere. What they do have is a trading record with the distributor they buy from every week, and that is what the model reads.
Built on verified invoices
Each invoice behind a score has been raised by the distributor and checked by a person. The history is the distributor’s own books, not a self-reported figure.
It declines to guess
Below four verified invoices between that shop and that distributor, the model returns no score at all rather than a low one. “A bad risk” and “we have not seen enough” are different findings, and a lender acts on them differently.
Every decision keeps its reasons
A priced loan carries the score it was priced on, feature by feature, frozen at the moment of the decision. A model published later cannot rewrite what was decided earlier.
Where the work happens
One API, and the surface each role actually has to hand. Everything is bilingual, and amounts are in FCFA.
WhatsApp, for the retailer
The request, the offer, the acceptance, the collection code and the reminders. The channel a shopkeeper already has open, on the handset they already own.
A web dashboard, for the office
Lending officers price and disburse, distributors verify collection codes, support works the call list, administrators own the platform and the score.
A mobile app, for the field
Where the work is not at a desk: a rep registering a shop in a market, capturing KYC on the camera, a signature on the glass.