Circle's volume grew 151%. Its revenue grew 7% .

The quarter stablecoin adoption stopped predicting revenue#

A fox reads the field before it reads the fence. That patient prowl is the posture a web3 marketing team needs this month, because stablecoin adoption and stablecoin revenue have just been caught walking in opposite directions, and the evidence was published by the company with the most to gain from them walking together.

Circle reported its second quarter 2026 results on 5 August 2026. Onchain transaction volume for the quarter was $14.8 trillion, up 151% year over year. Total revenue and reserve income was $701 million, up 7% year over year. Both figures sit in the same release, per Circle's second quarter 2026 results. Set them side by side and the tidy story about stablecoin adoption comes apart in your hands.

That gap is not a scandal, it is a structure. It is also the clearest lesson in vanity metrics available to the category this year, and the numbers are large enough that nobody can pretend they are noise. Almost every crypto adoption deck in circulation is built on a volume chart. Circle's own accounts show that the volume chart and the money chart stopped being the same chart. Anyone still selling stablecoin adoption on throughput alone is now selling against the issuer's own filing.

Seven growth rates, set side by side#

Ranked together, the quarter's growth rates make the point faster than any paragraph can. The fastest movers measure throughput and float. The slowest movers are the lines that actually pay for the business, and the distance between the two ends of that list is the whole argument.

Read the bars the way a fox reads a hedgerow: the traffic is real, it simply is not walking down the path the marketing assumed. Stablecoin adoption is genuinely accelerating. The revenue attached to it is doing something far calmer, and a brand that sells the first number while its own accounts report the second is building a brittle brief that breaks on contact with a finance team.

There is a version of this piece that scolds the category, and it would be the wrong piece. The volume figure is not a fabrication, it is a faithful count of a real thing. The trouble is placement. It is being used as proof of a commercial claim it was never built to carry, and stablecoin adoption deserves a better argument than a number borrowed from the wrong column.

Revenue is a rate, not a toll on transactions#

Here is the mechanism, because the mechanism is the whole article. Circle reported reserve income of $668 million for the quarter, up 5% year over year. The company's own explanation, paraphrased from its release, is that reserve income growth came primarily from 25% growth in average USDC in circulation, partially offset by a 66 basis point decline in the reserve return rate, per Circle's results release.

Sit with those two figures. Average balances rose by a quarter. The reserve return rate fell by 66 basis points. What came out of that collision was 5% growth in the biggest revenue line in the business. A quarter more float, a fraction more income. That collision is the honest shape of stablecoin adoption economics, and it fits on a single slide.

That is what it means to earn on reserves rather than on transfers. Float first, flow second. USDC is a reserve-backed digital dollar, and the reserve composition and attestation material sits on Circle's transparency page. That page, and the rest of the Circle corporate site, is exactly where a curious buyer goes the moment your deck claims that rising volume means rising revenue.

So the 151% is real, the 7% is real, and neither number is lying. They measure different things. Onchain transaction volume counts how much moved. Reserve income counts how much sat still, and at what rate. A stablecoin marketing story that conflates the two is not exaggerating, it is misclassifying, which is worse, because misclassification is the kind of error a single analyst question dismantles in public. Stablecoin adoption charts and revenue charts answer two different questions, and a deck has to say which one it is answering.

The cost side settles the point. Distribution and transaction costs were $412 million, up 1% year over year. Adjusted EBITDA was $143 million, up 8%. Net income from continuing operations was $48 million, up $530 million year over year, which is the one line in the release that genuinely deserves a celebration and will not get one here, because it is a swing rather than a trend.

Ranked in dollars, the business becomes legible in five bars. Everything a marketing team wants to celebrate happens above that structure, never instead of it, and any claim that ignores the structure is a claim the structure will eventually correct. That is the quiet cost of a stablecoin adoption story told in the wrong unit: it does not fail loudly, it simply stops being believed.

The number a web3 brand should actually quote#

If the volume chart is the wrong hero, something has to replace it, and the replacement is sitting three lines further down the same release. Circle reported 175 financial institutions enrolled in Circle Payments Network at quarter end, up 29% quarter over quarter, per Circle. A stablecoin adoption claim built on that line is a claim about who has signed up, and a buyer can check it.

That number tracks distribution rather than throughput, which is precisely why it is the better claim for a business-to-business brand. An institution enrolling is a distribution decision: it is made by a committee, minuted, and awkward to reverse. A trillion dollars of transfers is an aggregate that can double because one participant changed how it batches. One is a relationship. The other is a rounding behaviour. Publish the relationship count and a stablecoin adoption claim survives the first hard question in the room. Stablecoin adoption measured as distribution is slower, smaller and considerably more persuasive.

The same logic explains the quarter's quietest strong number. Other revenue was $34 million, up 41% year over year, driven by subscription and services growth, per Circle's release. It is a small fraction of the size of reserve income, it is growing at several times the rate, and it does not care what happens to the return on reserves next quarter. Circle Payments Network is the product that story sells.

Defining that metric takes an afternoon and saves a year. Write down what counts as enrolled, who signs the count off, and the date it is taken on. Then leave the definition alone. The value of a distribution metric sits entirely in its stability, and a definition that drifts to flatter a quarter is worth less than no number at all.

The neighbourhood agrees on which counts carry weight. The Block reported on 5 August 2026 that more than 85 companies have joined Mastercard's crypto partner programme, which is another count of enrolled counterparties rather than a throughput total. When two of the largest players in the category both choose to publish a participant count, that is a signal about which number survives scrutiny.

The second version is duller and it is the one a chief financial officer can repeat without checking anything first. That is the entire test, and most decks fail it in the opening slide.

What regulators and ad platforms let you claim#

None of this happens in open country. A stablecoin marketing claim in the European Union sits inside the Markets in Crypto-Assets framework, whose scope and timeline ESMA maintains, and a claim in the United Kingdom sits inside the financial promotions regime the Financial Conduct Authority sets out for cryptoasset firms. Both regimes care about the same thing your finance team cares about: whether the claim is fair and not misleading. A stablecoin adoption claim is a financial claim in both jurisdictions, whether or not a lawyer wrote it.

The advertising layer opened further this summer, on conditions. Google announced on 22 July 2026 that from August it would allow the promotion of cryptocurrency exchanges, software wallets and hardware wallets in three more European Economic Area markets, Iceland, Liechtenstein and Norway, provided the advertiser is licensed as a Crypto-Asset Service Provider under MiCA by a relevant national competent authority. Google states that it will give a seven-day warning before suspending an account, per Google Ads policy help.

Read that as a deadline rather than a detail. Seven days is generous by platform standards and brutal by planning standards, because a licence application is not a seven-day project. Where a paid programme touches those markets, the licence status of the entity on the account has become a marketing dependency rather than a legal footnote, and it belongs on the same dashboard as spend. Licence status is now part of what a stablecoin adoption claim costs to make in Europe. The same discipline applies to the copy: a stablecoin adoption figure with no period, no source and no unit is exactly the promotion both regimes were written to catch.

The wider institutional conversation is running the same way. The Bank for International Settlements is the standing forum where central banks work through what a settlement asset should be, and its output is the context that every serious stablecoin adoption claim is eventually read against. Writing for that reader costs nothing and buys years.

Writing the growth story your numbers can defend#

So what does a team actually do on Monday. The work is small, unglamorous and mostly editorial, which is exactly why it keeps getting skipped in favour of a bigger chart. Every step below exists to make a stablecoin adoption claim a finance team will sign.

Pick the metric that moves the money, publish it in the same place you publish everything else, and let the volume chart live in the appendix where it belongs. That is the whole discipline, and it is the same one we apply in web3 marketing and in FinTech marketing, where the distance between a usage number and a revenue number is the oldest trap in the category.

It is a positioning problem before it is a reporting problem, which is why the fix starts in brand strategy and lands in content marketing. The claim you can defend is a narrower claim, and a narrower claim is a sharper one. Nobody has ever lost a pitch for being specific.

The AI layer raises the stakes again. Answer engines lift self-contained sentences, so a page announcing that stablecoin adoption is exploding will be quoted saying exactly that, with your name attached, long after the quarter has turned. Building pages that get quoted accurately is the point of SEO and GEO work, and it is cheaper than issuing a correction.

We have written the adjacent versions of this before, on stablecoins, trust charters and the compliance story and on legitimacy marketing after the CLARITY Act. The pattern repeats without much variation: the number that is easiest to say is rarely the number that is easiest to defend.

A fox does not chase the loudest rustle in the undergrowth. It works out which way the wind is walking, then waits at the quiet end of the trail. Stablecoin adoption is real, rising and worth building for, and the revenue attached to it moves on a different clock. A brand that says so out loud will still be trusted when the rate turns, which it will.

If you want the growth story built from the lines that survive a board briefing, that is the work folkfox does, and the conversation starts with your numbers rather than ours.

Frequently asked questions#