Visa rewrites stablecoin volume figures without disclosing transaction data shifts

• Visa’s adjusted stablecoin transaction count fell by less than 2% after the September 18 refresh, far less than the volume measure.
• The underlying Allium identity set grew from about 15 million labeled addresses to roughly 600 million in the new full set.
• Visa’s live transaction methodology page still cites “over 3 million” labeled addresses, a figure unreconciled with the 600 million changelog number.

• <2% drop in adjusted transaction count versus the volume-measure decline
• 600M labeled addresses now, up from about 15 million previously
• 60% of 2025 Ethereum stablecoin transfers sat inside complex transactions, per BIS

Visa’s September 18 data refresh lowered its adjusted stablecoin volume figure sharply, while its adjusted transaction count moved by under 2%, according to CryptoSlate’s analysis of the release. The divergence stems from reclassification, not a documented change in payment behavior. Visa’s changelog attributes the reset to a fuller set of address labels and revised filters, rather than any shift in underlying blockchain activity.

Allium’s Labeled Address Set Jumps From 15 Million to 600 Million

Visa says the definition of adjusted volume itself has not changed. The metric still aims to exclude labeled exchanges, contracts, bots, bridges, other infrastructure, and minting and burning activity. What changed is coverage: with far more addresses now identified, a larger share of recorded transfers falls outside the adjusted total than before.

The refresh also added heuristics for short-term routing. It changed how Visa separates organic activity from payment-labeled activity.

Visa’s own example illustrates the mechanism at scale. An automated program on Solana cycled the same stablecoins through thousands of throwaway wallets, moving very large sums across relatively few transactions.

That pass-through pattern is now excluded from adjusted volume across multiple chains, though Visa does not say whether the same program ran on other networks, or quantify how much it moved on Solana specifically.

Transaction Methodology Page Still Cites 3 Million Addresses, Not 600 Million

The mismatch between the dated changelog and Visa’s live transaction methodology page is unresolved. That page still describes “over 3 million” labeled addresses, less than 1% of the changelog’s new 600 million figure. Visa’s public text does not clarify whether the 3 million figure refers to a narrower subset or simply has not been updated, so it cannot serve as the pre-reset baseline.

Visa does not publish matched pre- and post-refresh adjusted volume totals in the changelog. The size of the value revision cannot be calculated from that disclosure alone.

The gap matters most in cross-chain comparisons. Visa’s published material offers no matched-window, same-definition results showing how much of the revision landed on Ethereum, Tron, Solana or any other network, a limitation relevant to institutions using the dashboard to weigh chain-level exposure.

BIS Study Finds 60% of Ethereum Transfer Events Sit Inside Complex Transactions

A separate Bank for International Settlements study of 2025 Ethereum activity involving USDT, USDC and PYUSD found that nearly 60% of transfer events occurred inside complex transactions bundling multiple token movements. Counting each event as a standalone payment overstates activity, though the finding applies only to that Ethereum sample, not to Tron, Solana or Visa’s broader adjusted total.

A related BIS study found the same stablecoin can serve different purposes by chain. Ethereum use ties more closely to smart-contract interactions, Tron more to holdings outside contracts.

Visa’s own “retail sized” bucket, covering adjusted transfers below $250, is not automatically a purchase or a distinct user either.

Even market size is measured differently: CryptoSlate listed USDT market capitalization at about $183.79 billion on September 26, a stock figure that cannot fill the gap left by Visa’s missing volume series, which measures value transferred over time rather than tokens outstanding at a moment.

Firms weighing on-chain settlement infrastructure face the same reconciliation problem Visa has yet to solve publicly.

The CCS read. We think institutions citing Visa’s dashboard in due-diligence memos should stop treating the adjusted volume line as a payments indicator until Visa reconciles its 3 million and 600 million address counts. A metric that can swing on a labeling update without a matching transaction-count move is a data-quality signal, not evidence that stablecoin settlement demand from banks or payment firms changed at all.

Visa has not said whether it will publish a reconciled address count or a matched, same-definition chain breakdown to close the gap between its two conflicting methodology pages. Until it does, comparisons of Ethereum, Tron and Solana stablecoin activity built on the September 18 reset remain unverifiable against any prior baseline.