For many veteran payment professionals in China, the old way of classifying the industry was familiar: internet payments, mobile phone payments, bank card acquiring, prepaid cards, and similar categories. People were used to judging what a payment company could do by looking at channels, product forms, and payment scenarios.
That later shifted as Chinese regulation began dividing the sector by the substance of the business into two buckets: stored-value account operation and payment transaction processing. The underlying test was simple: do you receive the payer’s prepaid funds or not?
For many incumbent payment institutions, that licensing shift did not immediately change day-to-day operations. A system does not get rebuilt overnight just because the name on the license changes.
Legally and from a regulatory standpoint, though, it marked a clear boundary. Processing money and holding money are not the same thing. One is about helping a customer complete a movement of funds. The other means customer money enters your system first.
That boundary becomes even more visible in cross-border payments.
Hong Kong has a standalone Stored Value Facility, or SVF, regime. In Singapore, the Payment Services Act breaks services down further into categories such as Account Issuance and E-money Issuance. Once a business starts building customer balances and carrying customer funds, the burden rises sharply across safeguarding, fund segregation, ledger requirements, anti-money laundering, sanctions, fraud controls, and bankruptcy remoteness.
That is why overseas payments make one thing increasingly obvious: remittance follows one logic, acquiring follows another, and payment processing follows another.
Once a company starts offering accounts, wallets, stored value, or multi-currency balances, the nature of the business changes. It is no longer only about moving money. It has started to hold money.
Accounts are often the heaviest layer inside the payments stack. The licensing burden is heavier. Compliance is heavier. Safeguarding obligations are heavier. Ledger standards are higher. System design is more complex. Operational responsibility expands.
That sets up the central paradox of the piece: if accounts are so difficult, why are the biggest payment companies pushing toward them?
Stripe has extended from payments into Financial Accounts, Cards, and Financing. Adyen has moved from payment processing into Business Accounts, Issuing, and Capital. Airwallex has expanded from cross-border payments and FX into Global Accounts, Multi-currency Balance, and Treasury.
The answer, the article says, is that payment and accounts solve entirely different problems. Payment is an event. Account is a state.
Payment addresses how money moves. An account determines where money sits before movement, who owns it after movement, how much remains, and where it can go next.
Put differently, payment helps a customer move money once. An account lets a provider stay involved in every movement that follows.
That is why the deeper a company goes into payments, the more it ends up chasing the hardest layer in the stack: the account.
Payment is an event. Account is a state.
From the user’s point of view, a payment may last only a few seconds: click, authorize, succeed, done. At the level of financial infrastructure, that is only one money movement. What matters over time is the resulting change in account and ledger state.
If A pays B $100, payment describes the arrow in the middle — the movement. What the financial system must maintain over time is the state on both ends of that arrow.
Who owns how much money? Is a balance available or pending? In what currency? Under which legal entity? Has settlement happened? When can the funds move again?
Those are account-and-ledger questions.
Payment can run over very different rails, but it still ends in a balance and ledger update. Cards require bookkeeping. Bank transfers do too. Wallets do too. Stablecoins also require a ledger record of balance and ownership.
Payment methods may keep evolving, but the financial system does not escape two basic questions: money ownership and ledger state.
For the past 20 years, the payments industry has been exceptionally good at optimizing the arrow itself. Faster. Cheaper. Better success rates. Smoother checkout. Smarter routing.
As payment infrastructure matures, the commercially more valuable question shifts. It is no longer only how money gets from A to B. It is who controls the accounts at A and B.
Why accounts create a deeper commercial relationship
Payment creates a transaction relationship. Account creates a financial relationship.
If the lens is a single transaction, payment is obviously important. Over a five- or ten-year customer life cycle, though, an account represents a very different level of relationship depth.
A large enterprise may use Stripe today, add Adyen tomorrow, and connect a local PSP after that. Multi-PSP setups are standard industry practice among larger merchants.
Different providers come with different pricing, authorization rates, local coverage, and risk appetite. Enterprises also use multiple channels to reduce single-point risk.
That gives payment a strong routing character. One transaction can move from provider A to provider B. A PSP can lose traffic on the very next transaction because of success rates, pricing, or routing strategy.
That means a pure payment relationship is less durable than it may seem.
But if an enterprise puts collections, balances, payouts, FX, corporate cards, supplier payments, financing, and treasury inside one system, the relationship changes in kind.
Replacing a payment gateway may be a technical integration project. Migrating accounts, balances, payments, cards, FX, and treasury is much closer to a finance infrastructure migration.
That is the article’s point in a line: payment is a transaction, account is a relationship.
So when a payment company moves into accounts, it is not just trying to sell one more product module. It is trying to change its place in the customer’s financial stack — from participant in a transaction to owner of a longer-term financial relationship.
Payment appears at one moment. Account covers the full life cycle of funds.
A pure payment provider is mainly present at the moment money moves. A company’s cash life cycle is much longer than that.
Before money comes in, there are questions about where to collect it, which rail to use, and in what currency it should arrive.
After it comes in, there are questions about where to hold it and whether to convert it immediately.
When funds move, there are questions about rail selection and routing.
After the transaction finishes, there are more questions: when to settle, whether to pay out or keep the balance, whether to pay suppliers, run payroll, use a corporate card, make an intercompany transfer, or move the funds into treasury management.
That stretches a payment into an entire cash chain:
- Collect
- Hold
- Convert
- Pay
- Spend
- Finance
This is why the product boundaries of leading payment firms now look increasingly similar. Payments lead to accounts. Accounts lead to FX. Then come cards, capital, and treasury.
The commercial logic is straightforward. Payment earns revenue from money movement. Accounts compete for the value created while funds remain in the system and through the next financial actions attached to those funds.
Once a payment ends, the processing fee is largely fixed. If money remains inside the account system, it can connect to FX, payouts, card spending, financing, treasury, and broader financial product distribution.
Those are not equally deep business models.
From transaction economics to balance economics
The article adds another layer that often gets overlooked in payment discussions: balance.
The author says they previously examined cross-border payment data for LianLian, Payoneer, and Wise in the first half of 2025. Alongside TPV, revenue, and take rate, they pulled one more metric: customer funds on hand.
Using a consistent methodology, the customer fund balances for the three firms were about $2.22 billion, $7 billion, and $25.3 billion.
Those balances are not the companies’ own money, and they do not imply those institutions are free to invest customer funds at will. Different jurisdictions, license types, safeguarding structures, and banking agreements impose very different rules on where customer funds can be kept, whether they can be invested, how yields are handled, and who receives them.
Even with those differences in mind, the data still points to a major commercial shift. When a payment company manages customer balances in the billions or tens of billions of dollars, it is no longer only processing transactions. It is managing balances.
And once the business enters balance, the economic model changes.
A pure processing model revolves around transaction economics: one transaction happens, one processing revenue event is generated, and that transaction ends.
An account model enters balance economics: a balance that remains in the system can connect to FX, payouts, cards, liquidity management, treasury, financing, yield products, and other economics formed around customer funds and banking partners.
That means accounts do not only add more product revenue lines. They change the center of gravity of the business.
In the author’s framing, payment economics revolve around transactions. Account economics begin to revolve around balances.
The move from transaction economics to balance economics may be one of the most underestimated upgrades in the business model of a payment company moving into accounts.
Payment sees transactions. Account starts to see the business.
PSPs already have a large amount of useful data: TPV, approval rates, refunds, chargebacks, ticket size, transaction frequency. That data says something about how a merchant is selling. It is payment data.
Once a provider goes deeper into accounts, the data changes. How much came in today? How much went out today? What is the real-time balance? What are the positions by currency? Which entity is showing a sustained net outflow? Is working capital getting tight? Is cash flow stable?
The dataset moves from payment data to cash-flow data.
That shift matters. Payment data tells you how a company is selling. Account data starts to tell you how that company is living.
Once transaction data and cash-flow data are linked, the upper layers can naturally extend into lending, working capital products, risk assessment, cash-flow forecasting, and treasury services.
That is why the article says payment sees the transaction, while account starts to see the business itself.
Accounts are what can become a financial operating system
If a company only uses your system for acquiring, you are a payment provider.
If that same company opens your system every day to check balances, manage multi-currency positions, execute FX, pay suppliers, issue corporate cards, manage expenses, obtain financing, and run treasury, your product has entered the finance team’s actual core workflow.
At that point, what is being sold is no longer just payment. It is a financial operating system.
The expansion of payment company product lines is not simply a matter of chasing whatever generates revenue. The deeper fight is over who becomes the default operating entry point for daily money management.
Banks have done accounts for centuries. What did fintech actually reinvent?
If accounts matter this much, banks have been doing them for centuries. So what has fintech really reinvented?
The article’s answer is that fintech has often not reinvented the bank account itself. It has reinvented account architecture and account experience.
What the industry calls an account today can refer to very different layers.
- A legal bank account is the formal bank account under law, with the bank carrying the related balance sheet, capital, liquidity, and regulatory responsibilities.
- A payment account or wallet records customer balances and rights within a payment system.
- A virtual account is often used for identification, collection, and reconciliation. It may have its own account number even when there is no separate legal bank account underneath.
- A ledger account records internally who owns what, how much is available, how much is pending, and when funds can settle.
In that sense, many fintechs are not recreating a bank balance sheet. They are abstracting an extremely complex underlying structure.
In the past, a global company might need separate bank accounts in the US, Europe, and the UK, different currency balances, FX accounts, corporate cards, and multiple payment systems.
Today, a fintech can package that into one global account, one multi-currency balance, one API suite, and one dashboard.
The customer sees a simple account layer. Underneath it may sit multiple banking partners, local clearing networks, SWIFT connectivity, different currencies, different legal entities, and different safeguarding structures.
That means fintech’s real contribution is the software-ization of complex financial infrastructure. It does not necessarily take the bank’s balance sheet. But it does begin to compete for interface, ledger, workflow, and most importantly, the customer relationship.
There is a less discussed reverse side to that, though. The account can be abstracted at the front end, but the dependence on banking underneath does not disappear.
A global account may look to the customer like one dashboard, one API, dozens of currencies, and dozens of countries. Under it may rely on a sponsor bank, safeguarding bank, settlement bank, local clearing bank, correspondent bank, and different banking partners in different markets.
That creates a practical issue. A fintech may own the account interface without owning the ultimate balance sheet.
If an underlying bank changes its risk appetite, exits an industry category, tightens a country corridor, raises compliance thresholds, or ends a bank-fintech partnership, the seemingly stable account layer above it may need rerouting, account migration, new KYC, or even a rebuilt flow of funds.
So the account business has its own paradox. Fintech can hide banking complexity, but it does not eliminate banking dependence.
As the article puts it, software can abstract banking complexity, but it cannot software-ize balance sheet risk.
A mature account platform therefore depends on more than APIs and dashboards. It also needs depth with banking partners, multi-bank redundancy, safeguarding structure, liquidity management, reconciliation, and the ability to move flows quickly when one bank exits.
Having the account relationship, in other words, is not the same as having account sovereignty.
A truly strong account infrastructure has to solve both sides at once: customer experience at the front end and banking resilience underneath.
Why cross-border payments are especially likely to move from payment to account
The same logic becomes clearer in cross-border payments.
Domestic payments often involve a relatively simple single-currency movement, such as CNY to CNY. The customer pays, the merchant receives renminbi, and the payment is over.
Global businesses do not operate that way.
A company may receive USD, EUR, GBP, JPY, SGD, and KRW at the same time. Those funds are not necessarily converted at once into the headquarters’ functional currency.
What the business has to manage is where the money is, which entity owns it, in which currency it sits, which funds should stay local, which can be netted, which entity is short on liquidity, when to convert, and when to repatriate.
That means the real complexity in cross-border payments is not only money movement. It is money position.
The article gives the example of a Chinese global company with EUR revenue in Europe, USD revenue in the US, and JPY revenue in Japan; suppliers mainly needing CNY; overseas advertising platforms needing USD; a Singapore team needing SGD; and multiple legal entities that also need working capital.
In that setup, the finance team is no longer asking only how to send out an international transfer. The daily questions become whether to convert EUR now, how much USD to keep, where natural hedges exist, where netting is possible, which entity needs a liquidity injection, when to execute FX, where a liquidity buffer is needed, and when to repatriate funds.
That is not simple payment. It is treasury.
This is why cross-border payment companies are more likely than single-currency PSPs to move toward accounts, and why the move is more urgent. Customers do not ultimately need just a payment rail. They need global money management infrastructure.
In cross-border payments, then, payment solves movement and account starts managing position.
From there, the path is almost natural:
Account → FX → Liquidity → Treasury
A mature cross-border payment company will find it hard to remain only a transfer business over the long run.
Banks are moving up. PSPs are moving down.
If the global payments industry of the past decade is viewed as a vertical value chain, two forces are moving toward each other. Banks are moving upward. PSPs are moving downward.
The author says this trend also came up in recent conversations with people at JPM.
Banks have traditionally been strongest at the bottom of the financial system: accounts, balance sheet, clearing, liquidity, and compliance. They control genuinely scarce infrastructure.
But banks also face a structural problem. They may have the balance sheet while drifting farther from the business scenario.
The systems enterprises open every day are often not internet banking portals. They are ERP systems, e-commerce back ends, SaaS tools, treasury management systems, PSP dashboards, and other enterprise software.
The customer’s business workflow sits at the top of the stack. The bank’s balance sheet sits at the bottom.
If a bank protects only the balance sheet while losing client interface and distribution, it risks being compressed into a backend financial utility.
That is why banks need to move upward into API banking, virtual accounts, embedded finance, ERP integration, treasury integration, and embedded receivables and payables. The aim is the same across all of them: get closer to the customer.
PSPs begin from the opposite side. They start close to the customer. They own the checkout, the payment experience, and the merchant relationship.
But once a payment company reaches scale, it runs into another reality. Distribution does not automatically create a deep financial relationship.
Processing margins remain under pressure. Gateways are replaceable. Payment methods are increasingly standardized. Large merchants use multiple PSPs. Transactions can be rerouted at any time.
If a PSP stays forever at the top layer, it risks becoming a processing layer that is priced, switched, and replaced with relative ease.
So PSPs also have to move — downward, into accounts, ledgers, FX, cards, treasury, and financing. The goal is the mirror image of the bank’s: get closer to the money itself.
This creates a symmetrical structure across the industry. Banks have the money and want more customer entry points. PSPs have customer entry points and want deeper money relationships.
The article says the two sides meet around three things: account, data, and flow.
Account determines who owns the long-term financial relationship. Data determines who truly understands the customer’s business. Flow determines who controls the path of funds and the distribution of downstream products such as FX, treasury, and financing.
That is why this is more than product strategy. In many cases it becomes a necessity.
If banks do not move upward, they risk losing distribution and client interface. If PSPs do not move downward, they remain in a highly competitive transaction-processing layer that can be routed away from and squeezed on margin.
Both sides are chasing the same things: stronger customer stickiness, wider profit pools, richer data, and a higher financial share of wallet.
That is why the product boundary between banks and PSPs is getting blurrier. The industry has not lost its division of labor. The two sides are moving from opposite ends of the value chain toward the most valuable middle position.
Banks want to move from balance sheet to relationship. PSPs want to move from transaction to relationship. The final contest is over the same prize: the primary financial relationship.
The real fight is not over an account. It is over the primary financial relationship.
Payment companies used to be judged mainly by TPV, take rate, authorization rate, payment method coverage, and the number of countries covered.
Those metrics still matter. The article argues that another question now matters just as much: how much of the customer’s financial relationship does the company actually control?
Traditional banks are strongest in deposit relationships. Real funds settle into the banking system, and banks hold the balance sheet.
PSPs first captured transaction relationships. Customers came to them when they needed to make a payment.
Account platforms are trying to capture something else: the operating relationship. How much came in today? How much balance is left? When should payments go out? Which currency should be converted? Where should funds stay? Which entity is short on liquidity?
If those daily money decisions become concentrated on one platform, that platform gets much closer to the customer’s true primary financial relationship.
This is also where a new layering is emerging. The legal banking relationship and the operating financial relationship no longer have to belong to the same institution.
A company’s underlying balance sheet, clearing, and liquidity may still come from traditional banks such as JPM, Citi, and HSBC. But the interface the finance team actually opens and uses each day may come from Stripe, Adyen, Airwallex, or another financial platform.
Banks may own the balance sheet. Fintechs may own the interface and workflow. Both matter. But the point where commercial value starts being redistributed is the main operating entry point for managing money.
Once a company controls that entry point, it is not just capturing the current payment. It is in position for the next FX trade, the next payout, the next card, the next financing product, and the next treasury decision.
That is the larger commercial value of the account.
The article is careful to add that this does not mean every PSP will become a bank.
The deeper a provider goes into accounts and treasury, the more it runs into the weight of traditional financial infrastructure: settlement, liquidity, credit, compliance, risk capital, and balance sheet. Those are not things a polished dashboard or a few APIs can create on their own.
The more likely outcome is a further restructuring of the value chain:
- Bank: Balance Sheet + Regulatory Trust + Liquidity
- Fintech: Technology + Ledger Architecture + Orchestration
- Payment Network: Money Movement Infrastructure
- Platform: Business Scenario + Distribution
The most valuable positions will increasingly sit at the connections among those capabilities. Who owns the account interface? Who sees the cash flow? Who controls the flow? Who decides where the next money movement goes?
That is why account, a concept that has existed for centuries, has returned to the center of competition in payments and fintech. Not because the market suddenly realized that an account is a good business by itself, but because the real contest has always been over control of the money relationship rather than the account number.
Will stablecoins make accounts disappear?
The article closes by pushing the question forward. If accounts are so important, will today’s account form remain in place forever? Not necessarily.
When people talk about accounts today, they usually mean bank accounts, payment accounts, virtual accounts, or internal ledgers.
But stablecoins, tokenized deposits, and programmable ledgers raise a deeper question: does money state have to be recorded inside a traditional bank ledger? Not necessarily.
The state of a traditional bank account sits on the bank’s core ledger. The state of a payment account sits on the payment institution’s ledger. The state of a stablecoin can sit on a blockchain ledger.
In the future, what a company experiences as an account may not always be a traditional account number. It could be a set of wallets, a tokenized deposit, a programmable balance, or even a money position governed by smart contracts.
That does not overturn the article’s earlier logic. It reinforces it.
Stablecoins may change the form in which an account exists, but they do not change the problem an account solves.
No matter whether money state is recorded in a bank core, a fintech ledger, or a blockchain, the system still has to answer the same questions: whose money is it, where is it now, who has the right to move it, and where can it go next?
So what may change is account form. What is less likely to disappear is account function. And what financial institutions will still compete for is the account relationship.
The article treats neobanks as another version of that shift. Neobanks did not reinvent money state. What they redesigned was account experience, interface, and customer relationship.
Some neobanks hold their own banking license, with the balance sheet and ledger under their control. Others rely on sponsor banks, EMIs, or banking-as-a-service providers. The user sees a complete account experience, while the legal account and balance sheet remain with a partner institution underneath.
That further illustrates the point: front-end account experience, legal account, ledger, and balance sheet can increasingly be split across different institutions. In the future, the account may no longer be a product handled end to end by one institution. It may become a set of financial capabilities that are unbundled and recombined.
But whatever the technical form and institutional boundary become, the competition itself does not really change. Whoever controls the account relationship gets closer to the customer’s next financial action.
Payment is the entry point. Account is the relationship.
The article ends on a simple claim. Accounts are heavier, harder, and more heavily regulated, yet the industry keeps moving toward them because payment solves the movement of money while accounts solve the management of money.
Payment is a transaction. Account is a relationship.
Whoever controls the account gets closer to the customer’s next payment, FX trade, financing choice, and treasury decision.
So the real value of an account does not lie in the account number itself. It lies in how naturally it sits next to the customer’s next financial action.