Payward Trades Adjusted Revenue for a Stricter Platform Margin Figure

The Headline Swap

Payward, the privately held parent of Kraken, has replaced the number it leads with. Its third-quarter highlights open with total platform margin (TPM) at $487 million. Until now, the company led with Adjusted Revenue, which it put at $508 million for the second quarter in August. The swap matters because it changes how the business looks.

Payward publishes selected quarterly figures rather than SEC filings, so its headline choices carry extra weight. The release keeps Adjusted Revenue in a footnote, "for comparison purposes only," and states that "management no longer uses Adjusted Revenue to evaluate our business results." My read: the company has picked the metric it wants people watching, and it isn't the bigger number.

That's not unusual on its own. Companies revise their preferred metrics all the time, and private companies have more room to do it because they aren't bound by the same disclosure rules as public ones. What matters here is direction. When a company drops the larger number from its headline and moves the old one into a footnote, the new figure is almost always the one that tells the story management prefers. Read the footnote first.

What the New Metric Takes Out

Both measures start with total revenue and subtract trading costs, along with gains or losses on trading. TPM goes further and subtracts staking rewards paid to clients, earned when client crypto helps secure proof-of-stake blockchains. Payward says the change means "everything we pass on to clients is treated the same way." That's a defensible accounting logic, but it also shrinks the headline.

The logic is easier to see with an analogy. Think of a retailer that sells goods and also passes a share of each sale back to suppliers. Reporting the gross figure makes the business look bigger, while reporting the net figure shows what the company keeps. Neither is a lie, but they answer different questions, and the one you pick tells readers what you want them to focus on. Payward's argument is that staking payouts are pass-through money, not revenue it earns, which is a reasonable position. The trouble is that the old headline and the new one can't both be the "real" number at the same time.

The recast figures show the effect. Second-quarter TPM now stands at $436 million, roughly $62 million, or about 12%, below the $498 million Adjusted Revenue the release lists for that quarter. That $498 million was itself revised down from the $508 million first reported in August. On the old measure, third-quarter Adjusted Revenue was $553 million, up 11% from the second quarter and down 15% from a year earlier. Put the two third-quarter measures side by side and the new headline sits about $66 million lower.

Notice what that revision history implies. Two adjustments have now moved the second-quarter figure, and both moved it down. That doesn't prove anything improper, and companies restate figures for legitimate reasons all the time. But a reader tracking the series has to keep re-basing their picture of the company, and that's exactly the kind of friction that makes a metric harder to trust.

Where the Revenue Comes From

Asset-based and access-based TPM, which cover what Payward earns on client balances and what clients and partners pay to use its products, made up 52% of the total, up from 46% a year earlier. In May, co-CEO Arjun Sethi said in a podcast interview that "the right way to read Payward in 2026 and beyond is not, hey, you're just a bigger crypto exchange." The mix shift fits that argument. Here's the thing, though: a higher share from balances and access fees is a different business than trading volume, and the numbers are starting to show it.

That shift has real consequences. Trading revenue tends to rise and fall with market activity, so a business built on it can look spectacular in a hot quarter and ordinary in a cold one. Revenue tied to balances and access fees usually behaves more steadily, because clients keep assets on the platform and keep paying for services whether or not they're trading that week. That's the appeal of the pitch. The catch is that steadier revenue is rarely the fastest-growing kind, so a company leaning into it may look slower on the headline even while the underlying base gets sturdier.

And yet the shift cuts both ways. A business that earns more from balances is also more exposed to what those balances are worth, and to whether clients keep them on the platform. The release gives you the mix, but it doesn't give you the full picture of what drives each part, and that's where the skepticism belongs.

Profitability Tells a Messier Story

Adjusted EBITDA, which excludes items such as stock-based pay, came in at $60 million. That's up from $23 million in the second quarter but down from $181 million a year earlier.

Adjusted EBITDA is the number investors tend to reach for when they want a rough profit proxy, and it's also the one that needs the most caveats. Stock-based pay is a real cost, even when it's settled in equity rather than cash, because it dilutes existing holders. Stripping it out makes the business look leaner than a strict accounting view would. Still, the sequential jump from $23 million to $60 million is a genuine improvement. The year-over-year drop from $181 million is the figure that gives it context, since a business can recover from a weak quarter and still sit well below where it was twelve months earlier.

Volume and Accounts

Payward's platforms handled $353 billion in transactions, up 22% from the second quarter but down 30% from the same period of 2025. Funded accounts reached 7.1 million, up 7% in three months and 37% in a year. That's a count of funded accounts, not unique clients, so it shouldn't be read as a user total.

Funded accounts is a useful number, but it's a blunt one. An account can be funded once and then sit idle for months, so 7.1 million tells you about reach rather than engagement. Platforms in this category often track active or trading accounts separately because that's closer to how much the business actually depends on its users. The excerpt here doesn't include that split, so you're left with a headline count and a question you'll have to answer somewhere else.

Transaction volume tells a similar story. A 22% rise from the prior quarter sounds healthy until you see that the same figure is down 30% from a year earlier. Crypto volumes are closely tied to market conditions, so swings like this are common in the category and rarely say much about the business on their own. The more useful comparison is how the revenue mix holds up when volume moves, which is exactly what the shift toward balance-based and access-based income is designed to answer.

TPM and Adjusted EBITDA are non-GAAP measures, and the release doesn't say either figure was audited. Payward confidentially submitted draft IPO paperwork to the SEC in November 2025. If a public listing ever happens, these definitions will face far closer scrutiny. Until then, the question to watch is whether TPM keeps growing while transaction volume stays volatile.

What This Means If You're Watching Payward

If you're tracking the company, start by reading the footnote, not the headline. The relationship between TPM and Adjusted Revenue tells you what management is comfortable emphasizing, and the gap between them tells you how much the definitions matter. A reader who ignores the footnote is working from the version the company chose to show.

The bigger question is what this change signals. Companies usually swap a lead metric when the old one stops flattering them, or when they want investors and the public to judge them on something they can defend more easily. Neither reading is automatically bad. But the timing, close to the public-listing groundwork Payward has already begun, means these definitions are likely to be tested later by people with more leverage to question them. Anyone building a view on this business should expect that the metric debate will get more intense, not less.

The quarter itself is mixed. Revenue mix is shifting in the direction the company has been arguing for, and profitability is recovering from a weak sequential base. Against that, the year-over-year declines in volume and profit are hard to wave away, and the restatements to earlier quarters give skeptics something to point at. This is more significant than the headline suggests, but not in the way management wants you to think. The number they're showing you is the one that makes the next few quarters easiest to explain, and that's worth keeping in mind every time a new release lands.