Payment giant Visa updated its Onchain Analytics database on September 18, lowering its reported adjusted stablecoin volume measure while its adjusted transaction count fell by less than 2%. The divergence stems from refined data filters and an expanded database rather than an actual drop in real-world payment activity.
Key Takeaways
- Visa lowered its adjusted stablecoin volume metrics following a September 18 data refresh, while transaction counts dropped less than 2%.
- The underlying Allium identity set grew from approximately 15 million labeled addresses to roughly 600 million.
- An automated program on Solana moved large stablecoin sums through thousands of throwaway wallets before being excluded under revised heuristics.
- A 2025 Bank for International Settlements (BIS) study revealed that nearly 60% of transfer events in an Ethereum sample occurred within complex transactions.
Refined Filtering Shrinks Reported Volume
According to the Visa Onchain Analytics changelog, the data adjustment is tied to a larger address label set provided by analytics firm Allium, which expanded from about 15 million labeled addresses to roughly 600 million. Visa noted that its core definition of adjusted volume remained unchanged, aiming to filter out exchanges, smart contracts, trading bots, cross-chain bridges, infrastructure, and token minting or burning events.
Because dollar volume weights transactions by value while transaction counts assign one unit per event, removing high-value non-payment activity dramatically impacts volume without significantly changing transaction counts. For instance, Visa cited an automated program on Solana that cycled stablecoins across thousands of throwaway wallets. This pass-through mechanism moved substantial capital in relatively few transactions and has now been excluded from adjusted volume figures across multiple blockchains.
Despite the changelog citing 600 million addresses, Visa's live transaction methodology text still references over 3 million labeled addresses, leaving it unclear whether the figure refers to a distinct subset or an un-updated baseline.
Distinguishing Transfers From Real Payments
Visa's methodology classifies transactions into distinct categories including DeFi, centralized exchange flows, investments, store of value, and payments. Even within its "retail-sized" category—which filters for transfers under $250—a small transfer does not necessarily signify a retail purchase or merchant settlement.
This gap between blockchain movement and economic usage is further highlighted by a BIS study analyzing 2025 Ethereum activity across Tether (USDT), USD Coin (USDC), and PayPal USD (PYUSD). The study found that nearly 60% of transfer events occurred within complex, multi-step transactions, demonstrating that single on-chain events often represent automated liquidity routing rather than individual consumer payments. This comes amid broader industry developments as Washington eyes global stablecoin initiatives while USDT market capitalization reached $183.79 billion on September 26.
Why It Matters
Raw blockchain metrics routinely confuse bot activity, automated liquidity provisioning, and complex smart contract routing with genuine consumer commerce. As major payment networks refine their data pipelines to filter out artificial activity, market participants gain a clearer picture of real economic utility. Moving forward, evaluating stablecoin adoption will require looking beyond high headline transaction volumes to isolate true end-user payments from underlying automated infrastructure.
