Everyone agrees Africa runs on stablecoins. Nobody was measuring it.
The numbers in circulation came from surveys and extrapolation, and they disagreed with each other by tens of billions of dollars. A gap that size decides whether the market is real. So we stopped arguing about the estimate and went and counted it.
AfriFlux maps each venue’s on-chain infrastructure and attributes flow wallet by wallet, venue by venue. Every figure we publish carries the method that produced it, and every one of them is free to cite.
Cited by This Week in Fintech, Stablekoin, Techcrier, Fintech Is Easy, TechCabal, Digital Today, Condia, SolanaFloor, WagMedia and Mariblock. See the coverage
Four rules we don’t break.
We count it
Every figure traces to an on-chain sweep of addresses we have verified and labelled. Nothing here is extrapolated from a survey, a funding announcement or a company's own disclosure. If we can't attribute it, we don't publish it.
The method travels with the number
Each measurement states what was counted and what was excluded, so a regulator or journalist can check how any number was made.
Corrections in public
A published figure is never silently edited. When one changes, the note carries a dated correction and the original stays legible.
The measurement doesn't bend
The method is fixed before we know what it says about anyone — including the venues we cover, and including whoever is paying us.
Built by three people.
What each of them builds is below. What each of them thinks is in Founder’s notes.
Mukhtar Yahaya
Founder · Data Analyst / Engineer
Builds the attribution pipeline — the wallet-by-wallet mapping that turns raw on-chain movement into venue-level flow, and the method statements that ship with each published view.
@maybemukhtar →
Moshood Alimi
Co-founder · Software Engineer
Builds the product surface — the dashboards, the developer platform and the Terminal that put the dataset in front of the people making decisions with it.
@rogueboykillua →
Olushola Yahaya
Co-founder · Data Scientist
Builds the models behind the classification engine — separating genuine retail flow from treasury movement, and the cohort and momentum work behind the Gradient.
@le_magnifiq →Intelligence Feed →
93 published measurements, each dated and citable.
Terminal →
Invite-only access to the full attributed dataset.
Press →
Logos, photos, company description and press contact.
Want to work with us? Get in touch.