Methodology

01What we measure

AfriFlux measures stablecoin deposits into African venues — exchanges, offramps, neobanks, payment platforms and prediction markets. A deposit is money arriving at an address a venue issued to a customer and controls.

We count each deposit once, at the moment it arrives on chain. We do not count trades, conversions, internal transfers, or anything that happens after a venue takes custody.

Settlement protocols and tokenised-equity issuance are measured separately and are not part of the market total. A protocol's volume settles into venues this page already counts, so adding it would count the same money twice.

02How addresses are identified

Venues issue deposit addresses to customers and sweep them into collection wallets. We identify those addresses structurally — from the onchain behaviour that brings an address under a venue's control and the sweep path that confirms it — rather than from labels, registries or third-party lists.

The specific approach differs by how a venue builds its infrastructure. Every attribution is confirmed against a venue's own collection wallets before it counts.

03Retail and non-retail

Not all stablecoin flow into a venue is retail. Liquidity operations, institutional settlement, market making and business payments move through the same rails as individual customers.

Where a venue's book contains both, we separate them by deposit size and frequency and report retail separately. Where a venue serves individuals only, there is nothing to separate. Where the two cannot be told apart on shared rails, we say so and publish total flow rather than a retail figure we cannot defend.

The split is a size band. It does not know what a business is — a company can deposit retail-sized amounts, and an individual can deposit more than a company.

04Metric definitions

What each figure on the platform means, and what it cannot tell you.

05Observed vs inferred

Directly observed figures carry high confidence. Anything modelled is labelled as such, on the metric and on the page where it appears.

06Limitations

Stated plainly. Trust is built by showing the edges of what the data can and cannot say.

07Questions or corrections

If a figure looks wrong — especially if it is your venue — tell us. We will re-derive it and publish a correction with its impact if we got it wrong.

Every change to what we measure, and every published figure that has moved, is recorded on the methodology updates.

Reach us at motolayahaya@gmail.com.