Visa cuts adjusted stablecoin volume; payments unmeasured

On Sept. 18, Visa updated its Onchain Analytics dataset. The changelog shows the labeled-address set expanded from roughly 15 million to about 600 million. The refresh lowered adjusted stablecoin dollar volume while the adjusted transfer count declined by less than 2%. Comparable pre- and post-refresh payment totals were not published.

The changelog states the Allium identity set grew and that the definition of adjusted volume remained focused on excluding labeled exchanges, smart contracts, bots, bridges, infrastructure and minting and burning. The update added heuristics to detect short-term routing and altered how organic activity is separated from payment-related activity.

Public documentation does not include side-by-side adjusted-volume totals for dates before and after Sept. 18, nor matched-window chain results for Ethereum, Tron, Solana or other networks. Without those same-date comparisons, the revision’s effect on network-level totals cannot be calculated from the disclosed material.

Dollar volume and transfer count are calculated differently. A transfer count treats each included event equally, while dollar volume weights transfers by value. Removing a small number of very large pass-through transfers can substantially reduce reported dollar volume while leaving the transfer count nearly unchanged.

The changelog gives an example of an automated program on Solana that cycled stablecoins through thousands of throwaway wallets. Visa’s filters now exclude that pass-through pattern from adjusted volume across multiple chains. The example does not quantify how much of the revision applied to Solana or to the market overall.

The published pages show inconsistent labeled-address totals. A live methodology page references over 3 million labeled addresses, which does not match the changelog’s roughly 600 million figure. There is no clarification in the documentation on whether the smaller number refers to a subset or on whether the live wording has not been updated.

Visa’s transaction methodology separates payments from DeFi, centralized-exchange flows, trading, store-of-value holdings and other categories. A transfer included in the adjusted series may not be classified as a payment. The platform defines a retail-sized bucket for adjusted transfers below $250; a small transfer is not automatically a merchant payment.

Research cited in related documentation notes that stablecoins can serve different uses on different blockchains. A 2025 study of Ethereum activity found that about 60% of transfer events occurred inside complex transactions, where a single on-chain transaction can contain multiple token movements and operations. Counting each transfer event as a standalone payment can misstate underlying economic activity.

Token market capitalization measures outstanding supply at a point in time, while Visa’s volume measures value moved over a period. The changelog indicates the revision removed more high-value pass-through activity from adjusted stablecoin volume while leaving the adjusted transfer count nearly intact. Comparable dated payment-category and adjusted-volume data using the same definitions would be required to determine whether payment use changed or whether one blockchain gained share; those data were not released with the Sept. 18 refresh.