How to Measure Stablecoin Payment Volume

To measure stablecoin payment volume, start from raw onchain transfer records and subtract every transfer that is not a payment: minting and burning, bridge locks and unlocks, exchange-internal rebalancing, and high-frequency bot loops that move the same dollar back and forth. What survives that screen is the number worth quoting. According to Allium's stablecoins dataset, the total adjusted onchain stablecoin circulating supply is $311.75B (Allium data, October 12, 2026), and that supply figure is a common starting point for people who want a volume number, which is the first mistake to avoid.

Supply tells you how many dollars exist. Volume tells you how many dollars moved. Payment volume tells you how many dollars moved to pay for something. Those are three different numbers, and the gap between them is where almost every inflated headline comes from.

Key takeaways

• Circulating supply is a stock, not a flow. A $311.75B adjusted supply figure (Allium data, October 12, 2026) says nothing about how much was paid in a day. Never answer a volume question with a supply number.
• Raw transfer count overstates payments by a wide margin. Mints, burns, bridge events, and exchange internal moves are transfers but not payments.
• Automated loops (arbitrage bots, MEV, wash cycles) can move the same dollar hundreds of times. A dollar that returns to its origin in minutes is not payment throughput.
• A credible payment figure is a filtered subset of gross transfer value, defined by what you excluded, so publish the exclusions alongside the number.
• Cross-chain totals inflate if bridged supply is double-counted. The same native/bridged distinction that fixes supply also matters when you aggregate volume across chains.

Why supply is the wrong starting line

Supply moves slowly. Mints and burns adjust it over days and weeks. Of the $311.75B adjusted total, USDT and USDC alone account for $259.62B, or 83.3% (Allium data, October 12, 2026). None of that tells you whether a single payment happened today. If someone quotes a supply figure in a sentence about "payments processed," the units are already wrong.

The adjusted figure is itself instructive. The raw circulating supply reads $336.44B, while the adjusted figure is $311.75B (Allium data, October 12, 2026). The difference nets out bridged and wrapped supply that would otherwise be counted twice: a dollar locked on Ethereum and represented as a wrapped token on another chain is one dollar of backing, not two. The same discipline carries straight into volume. If you sum transfer value across chains without accounting for bridged representations, you count the same movement more than once.

The transfer stack, from gross to payment

Every transfer on a stablecoin contract looks similar at the ledger level: a sender, a recipient, an amount. The work of measuring payments is classifying each one. Here is the subtraction, laid out as a filtered funnel on a hypothetical day of gross transfer value.

The numbers above are illustrative, not a measured day. The point is the shape: the largest deductions are usually the automated and internal flows, and the surviving payment figure is a minority of gross transfer value. Anyone quoting the top of this funnel as "payment volume" is off by multiples.

Screening out the automated flows

The single hardest category to filter is high-frequency programmatic activity. A trading bot moving a stablecoin between two addresses it controls generates real transfers with real amounts, but no economic payment occurs. Practical signals that a transfer belongs in this bucket:

• Round-trip behavior: value that returns to its origin address within a short window.
• Counterparty concentration: a small set of addresses transacting with each other at high frequency.
• Contract interaction: transfers that are legs of a DEX swap, a lending action, or an arbitrage route rather than standalone sends.
• Timing regularity: machine-cadence intervals that no human or invoice cycle would produce.

A useful framing here is to separate gross transfer value from what some teams call attributed payment value, where each surviving transfer is tied to a real sender-recipient relationship. The distinction between raw volume and real activity is the whole ballgame, and stablecoin attribution is the method increasingly used to do it at scale.

A worked example: one dollar, one hundred transfers

Suppose one address runs an arbitrage loop that moves $1,000 of USDC between two venues 100 times in an hour.

• Naive transfer-value method: 100 transfers x $1,000 = $100,000 of "volume."
• Payment reality: zero payments. No goods, no services, no settled obligation. The same $1,000 never left the operator's control.

Scale that behavior across the market and you see why gross transfer value and payment volume diverge by an order of magnitude. The correct payment contribution of that address for the hour is $0, not $100,000.

Don't mistake a yield wrapper for a problem

One false positive worth naming: price. Most major fiat-backed stablecoins trade at par. USDT and USDC both read 1.0000, DAI reads 1.0003 (Allium data, October 12, 2026). But sUSDS reads 1.1127, above a dollar. That is not a depeg and not a measurement error. sUSDS is a yield-bearing wrapper designed to drift up in value, so its price above par is the product working as intended. If your screen flags it as anomalous, your screen is miscalibrated.

The field-level problem under all of this

Every exclusion above requires classifying a transfer, and that classification needs consistent fields. To decide whether a given transfer is a mint, a bridge leg, an exchange internal move, or a genuine payment, the same record has to resolve to the same shape on every chain: asset, issuer, sender, recipient, amount, USD value at transaction time, transaction type, and whether the token is native or a bridged representation. On Tron the transfer looks different at the raw level than it does on Ethereum, Solana, or Base, yet a payment figure that aggregates across them needs identical semantics. Allium ingests raw data from 150+ blockchains and normalizes these records into a standardized stablecoins dataset, including the native-versus-bridged flag that keeps a cross-chain total from counting the same dollar twice. That native/bridged split is concrete: $282.33B native against $29.42B bridged in the adjusted supply figure (Allium data, October 12, 2026). Bridged balances are flagged and reported separately rather than treated as independent dollars, and the equivalent care is what a defensible cross-chain payment number requires.

How to publish a number you can defend

Because "payment volume" has no single canonical definition, the credibility of your figure lives in its exclusions. State them explicitly:

• Which chains and which stablecoins you included.
• That you removed mints, burns, and bridge events.
• That you removed exchange-internal and known programmatic flows, and the signals you used.
• Whether you deduplicated bridged representations across chains.
• The date range and whether figures are gross transfer value or filtered payment value.

A reader can disagree with your filters, but they can only do that if you publish them. A payment number without a methodology is a transfer-count headline wearing a different label. For the reference architecture behind these flows, the stablecoin payment stack breakdown covers where each category of transfer originates.

Frequently asked questions

Is stablecoin circulating supply the same as payment volume?

No. Circulating supply is a stock, the total dollars of a stablecoin that exist at a point in time. Payment volume is a flow, the value that moved as payments over a period. Allium's adjusted supply figure was $311.75B (Allium data, October 12, 2026), and that number says nothing about how much was paid in any given day. Answering a volume question with a supply figure is a units error.

Why is the adjusted supply figure lower than the raw figure?

The raw figure ($336.44B, Allium data, October 12, 2026) counts bridged and wrapped representations alongside the native supply that backs them. The adjusted figure ($311.75B) nets out that double-counting by flagging bridged balances using a native-versus-bridged indicator and reporting them separately, so a dollar locked on one chain and wrapped on another counts once, not twice.

What transfers should I exclude to get a payment number?

At minimum: minting and burning (issuance and redemption, not payments), bridge lock/unlock/wrap events (collateral relocating), exchange-internal rebalancing (same beneficial owner moving funds), and high-frequency programmatic loops from bots, MEV, and wash activity. What remains after those exclusions is your candidate payment value, and you should publish the exclusions alongside it.

How do bot loops inflate stablecoin volume?

An automated loop can move the same dollar between addresses it controls hundreds of times. A $1,000 loop run 100 times produces $100,000 of naive transfer value but zero payments, because no goods or obligations were settled and the funds never left the operator's control. Round-trip timing, counterparty concentration, and contract-interaction patterns help identify these flows for exclusion.

Does a stablecoin trading above $1.00 mean something is broken?

Not necessarily. Most fiat-backed stablecoins trade at par (USDT and USDC both read 1.0000, Allium data, October 12, 2026). Some tokens, like sUSDS at 1.1127, are yield-bearing wrappers designed to appreciate over time, so a price above a dollar reflects accrued yield rather than a depeg. A measurement screen that flags these as anomalies is miscalibrated.

How should I aggregate payment volume across chains without double-counting?

Use a native-versus-bridged indicator the same way you would for supply. A bridged token is a claim on collateral locked elsewhere, not an independent dollar, so summing transfer value across chains without that flag counts the same movement more than once. Of the $311.75B adjusted supply, $282.33B is native and $29.42B is bridged (Allium data, October 12, 2026), and bridged balances are reported separately rather than treated as new dollars.

Interested in learning more about Allium’s stablecoin data? Speak to someone on the team.

Allium provides onchain data infrastructure. Companies named in this article may be Allium customers, prospects or commercial counterparties. The authors and other Allium personnel may hold positions in the assets discussed. This article is informational only and is not investment, legal or tax advice. Data and information last reviewed: October 12, 2026.