Visa refreshed its stablecoin metrics on September 18, reducing its adjusted volume measure while its adjusted transaction count fell by less than 2%. The divergence between these two measures reflects a change in how on-chain activity is classified rather than evidence of a shift in actual payment behavior.
The revision stemmed from an expanded address-labeling dataset. Visa's underlying identity set grew from approximately 15 million labeled addresses to roughly 600 million, allowing the company to better identify and exclude infrastructure activity, exchanges, bots, bridges, and minting-and-burning operations from its adjusted figures. The refresh also introduced new heuristics for short-term routing and updated how organic and payment activity are identified.
Why Volume Fell While Transaction Counts Held Steady
Transfer counts and dollar volumes can move in different directions because each transaction receives equal weight in count figures, regardless of value. Removing even a small number of high-value transfers can substantially reduce volume while barely affecting transaction counts.
Visa cited a concrete example: an automated program on Solana cycled large stablecoin amounts through thousands of throwaway wallets, generating many transactions but concentrating value in few transfers. This pattern is now excluded from adjusted volume across multiple chains. The mechanism is clear, though the company did not quantify how much of the overall revision this example represents or whether similar patterns appeared on other networks.
Classification Changes Don't Measure Payment Shifts
The recorded transfers themselves remained on-chain; only their classification changed. Visa's public disclosures do not provide comparable pre- and post-refresh adjusted volume totals, making it impossible to calculate the size of the revision. The company also did not publish matched-window, same-definition results showing how much of the change affected individual networks like Ethereum, Tron, or Solana.
A transfer surviving the revised filters is not automatically a payment. Visa's methodology separates payments from decentralized finance, centralized-exchange flows, investment activity, store-of-value holdings, and infrastructure transactions. Adjusted activity includes transfers below $250, but small transfers are not necessarily distinct purchases or merchant settlements.
Research from the Bank for International Settlements highlighted this distinction: nearly 60% of Ethereum transfer events involving USDT, USDC, and PYUSD occurred within complex transactions involving multiple token movements and financial operations. Counting each transfer event as a standalone payment can misstate underlying activity.
What Remains Unclear
Establishing whether actual stablecoin payment use changed, or whether one blockchain gained relative ground, would require comparable adjusted-volume and payment-category data under consistent definitions. The September refresh does not provide that level of detail publicly.
