Yuki Liu said she has decided to stop building in Web3 payments after spending the past six months moving from an observer of the sector to a participant working inside it.
In an essay published by MarsBit and attributed to Yuki Liu of Stablehunter/Money in Motion, she said the decision was not a retreat after failure. She described it as a shift in judgment made only after entering the business directly. Over that period, she traveled to Yiwu, Shuibei, Putian and Mexico to see how payment activity described in industry reports actually worked on the ground. She also built a Web3 payments MVP, handled accounts and worked on a crypto collection tool, trying to push an imagined path from the first step to the last.
The deeper she went, the clearer one point became, she wrote: payments is not a sector where a team wins simply by making a better product. What matters, in her view, is banking relationships, licensing, capital efficiency and the ability to control risk over a long period.
Liu argued that many payment businesses that look profitable are not really earning a premium on capability. They are earning a premium on risk, and have simply not yet run into trouble. The real test for a payment company, she wrote, is not how much it has made, but whether it can survive when risk becomes visible.
She said the article was not written to dismiss the sector. Her aim, she wrote, was to strip away the filter and lay out the structure more plainly so that others entering the field could make clearer judgments. She also noted that a few weeks ago she recorded a podcast on the same topic with former Kun Global VP Robert, Nayuta Capital CEO, and former Didi Financial CEO Alex.
Why she entered Web3 payments
Liu described herself as a repeat founder who ended a startup project that had run for several years. After closing that company, she took some time off and tried to think more carefully about where to put her energy next.
Half a year ago, a friend invited her to Hong Kong to explore a Web3 payments startup. At the time, she said, she was not familiar with Web3 itself and did not really know the payments industry either. At a macro level, though, it looked like a large market still in an upward cycle, and one where Web3 might also connect with AI.
In previous ventures, she had worked on cross-border business and also on platforms and software tied to remote employment. Those experiences kept bringing her back to the same fact: a business can go global quickly, but money still moves slowly. Settlement delays, fragmented routes, opaque costs and uncontrollable payment cycles may be tolerable when a company is small, she wrote, but those issues do not get solved by management once the business scales. They become larger. Money does not move as freely as information, and that acts as a hidden ceiling on many global businesses.
Against that backdrop, she said that when she started looking systematically at how Web3 payments were being used in clearing and settlement, what she saw was not an abstract technology story. It looked like a tool that could act directly on those pain points: faster settlement, more transparency and near round-the-clock clearing capacity.
At the time, she judged it to be a Day 1 Global direction that appeared to solve a real problem. She wrote that she did not enter because of Web3 itself, but because in the specific case of payments, it seemed to offer a better structure, at least logically, for attacking long-standing friction points.
Looking back, she now thinks she had accepted a premise that reality later challenged again and again: if clearing and settlement become efficient enough, payments will naturally move on-chain. In simplified form, she said, the idea became an intuition that payments are just transaction matching, and once the workflow runs, cash flow can be assembled by hand.
Because she did not know either Web3 or payments well enough at the start, she decided to spend three months going into the industry directly, understanding the structure first, and only then deciding what to build and where to stand.
What payments really compete on
When she arrived in Hong Kong, the initial plan was simple. Liu said the team wanted to lean on some resources and relationships already available through friends, start from OTC or relatively simple payout and collection scenarios, get cash flow moving first and then decide what to do next based on real demand. She was not there to observe from a distance, she wrote. She wanted to see whether something workable could be built first and the direction refined later through actual business.
Very quickly, though, the external environment accelerated. Liu said that in May, the U.S. passed the GENIUS Act and the sector was almost ignited overnight. Capital, projects and founders rushed in, and Web3 payments shifted from a niche infrastructure topic to a heavily discussed opportunity. That might have looked positive from the outside, she wrote, but for a team that had only just entered the field, the sudden excitement was not necessarily a good thing.
Moments that are noisy, mixed and full of fast-forming consensus often hide the real problems, in her view. Large internet companies, financial institutions, banks, traditional Web2 payment companies and Web3-native teams all started entering the space. Everyone talked about opportunity, she wrote, but few talked about structure. That made her more convinced she needed to stay close to the ground.
The view from reports did not match the front line
Once she started fieldwork, her first move was not to keep refining the product plan. She wanted to see who was actually using Web3 payments, why they were using them and where they were using them. She began with Yiwu, one of the places most frequently cited in industry reports.
In research notes and public discussions, Yiwu is often held up as a representative case showing Web3 payments already operating at scale. Liu said that what she saw in person looked very different. Stablecoins did appear in transactions, but usage was more scattered, more relationship-driven and more hidden behind the scenes.
It did not resemble the kind of standardized, productized settlement method often described in reports. Many transactions were not happening because stablecoins were obviously the most efficient option. She later went to Shuibei, Putian and Mexico, and also looked at penetration in places such as Africa and Argentina. She said the picture was not materially different.
Her conclusion was not that Web3 payments do not exist. It was that they have not formed a stable main route that can be scaled up reliably. More often, they function as a patch inserted into an existing system. Real penetration, she argued, does not match the level of heat seen in reports, communities and online discussion.
Those conversations also changed her frame. She began to move away from asking whether a product could be built and toward asking what the industry structure actually was. The incremental market for stablecoins, she wrote, may not sit inside crypto itself. It may sit in Web2 business scenarios that already exist but have been dragged down for years by traditional clearing and settlement systems. That was less a narrative rotation than a slow fintech upgrade. At the same time, a tougher question appeared: if real usage is this fragmented, can a productized path really hold up?
Once the team started building applications, every problem led back to rails
From July to September, Liu said, she continued field research while systematically engaging potential customers. These included HR companies, insurance, travel, MCN agencies, service trade businesses, cross-border operators and gaming companies. Their needs varied, but the core request was remarkably similar: money should move faster, cheaper and more reliably.
Payroll, task settlement and B2B payments all looked like sensible stablecoin use cases on paper. The team initially believed the application layer could be an entry point. But one condition quickly became impossible to avoid: any such business needs a stable, compliant and sustainable fiat-to-digital and digital-to-fiat channel.
The team began working with several service providers in the market that looked promising. After actual use, she wrote, it was hard to call any of those channels reliably durable. To meet business needs, the team even tried to patch the rails itself. Only then did it become clear that this was not a product problem at all. It was an infrastructure problem.
Bank relationships, licensing structure, KYB/KYC compliance, risk control, limit management and communication with regulators all sit inside the channel layer. In Liu’s account, that layer depends on accumulated credibility, experience and capital built over a long period. Those are not capabilities a small internet-background team can make up quickly.
That was the point, she wrote, when she first fully understood that payments is not a business where product quality alone wins.
What looks like profit may only be risk premium
One line stayed with her through the process: in payments, it is not about how much money you make, but how much money you can spend. Liu used that line to argue that many Web3 payment paths that appear to be working are not earning a premium on execution. They are earning a premium on risk.
What makes the situation more dangerous, she wrote, is that many participants do not know what risk they are carrying or where it sits. It could be counterparty compliance. It could be a mismatch in the structure of the capital pool. It could be delayed risk-control rules. It could be a gray zone in regulatory interpretation.
If the business works only because nothing has gone wrong yet, then it is not a structure that can be expanded with confidence, she wrote.
Payments as a water-flow business
Liu said she gradually came to view payments through a simpler lens. At its core, she wrote, payments is a business of water flow. Whoever controls the waterways can make money. The larger the flow through the tap, the larger the earning space. If water passes your doorstep, you can take a cut. From a distance, that can look almost effortless.
But that is also why the business is difficult. Not every company standing by the water actually makes money. The firms that sustain profits over time are usually those with strong control over volume, pressure, backflow, contamination and leakage.
How much water a company can take on depends on how much risk it can bear. How long the flow can keep running depends on its tolerance within compliance, risk management and regulation. Many routes that appear to have a large flow, she argued, are simply routes that no one has shut off yet.
That was the process through which she developed what she called a more complicated but more realistic respect for the payments business. Its appeal, in her description, is not that someone launched another new product. It is that the business tells you, very honestly, which industries are actually making money in the real world and which are only loud. If you stand on the waterway, you can see where capital is really going instead of watching public relations from the outside.
A good business, but not one her team is best equipped to build
By that stage, Liu said she had to face a judgment that was not easy for a founder but was necessary: payments is a good business, but not the kind of business her team could do best. She framed that not as a rejection of the direction but as respect for the team’s resource endowment.
The capabilities that matter in payments, she wrote, are not rapid trial and error or constant product iteration. They are long-term banking relationships, a sustainable compliance framework, mature risk controls and credibility formed through repeated interaction with regulators. Those are not assets a team can gain in the near term through effort or intelligence alone. They look more like industry-level assets that form gradually in specific kinds of teams and specific windows of time.
Once she saw payments as a water-flow business, she also became more certain that what decides whether a team can stand on the waterway for the long run is not desire. It is whether the team already holds the structure capable of taking that pressure.
Under those conditions, continuing to push forward no longer looked like a rational investment for her team. It looked more like using time and luck to fight against an industry structure that was not on their side. That conclusion led her to her next decision.
Still constructive on payments, but clearer on where the battle is
Liu stressed that stepping away from Web3 payments does not mean she is bearish on the sector. Quite the opposite, she wrote: over the past six months she has become more convinced that the payments industry still offers large structural opportunities.
What changed is her understanding of those opportunities. Once broken down, they point to a business with a longer cycle, heavier structure and higher resource demands than many founders first assume. The opportunity is real, she wrote, but it is not evenly distributed beneath every startup team.
The upside is long-term rebuilding, not a short-term windfall
Viewed over a longer horizon, cross-border payments are not a question of whether they will break out. They are part of an infrastructure rebuilding process already under way, in her telling. Supply chains continue to spread globally, cross-border trade in services keeps growing and distributed teams are collaborating faster. Together, those trends keep magnifying the friction inside traditional clearing and settlement systems.
In that setting, the value of Web3 payments is not simply that they are cheaper, she wrote. It lies in three areas:
- a meaningful improvement in turnover efficiency
- greater transparency in settlement paths
- unified settlement across currency zones and regulatory zones
She described that as a structural improvement rather than a tactical optimization. For that reason, she said, it belongs to a project measured on a ten-year scale, not a market that can be unlocked with a burst of product execution.
The hard part is not collecting money but building fund systems inside marketplaces
After enough exposure to real operating scenarios, Liu said she became increasingly certain that the core difficulty in payments is no longer the act of taking payment itself. In marketplace settings especially, payments are not a stand-alone module. They are a full ecological capital system.
Buyers, sellers, platforms, logistics providers, streamers, riders, tax obligations, frozen accounts and subsidy accounts all constrain one another on the same money rail. In that system, the true barrier is not the payment interface. It is:
- escrow and freezing mechanisms
- split settlement and payment-cycle design
- risk control and anti-fraud capability
- cross-regional compliance and regulatory obligations
Once such systems become stable, she wrote, they naturally gain room to extend into broader financial capabilities. But they also demand deep capital strength, a mature risk framework and patience over a long period.
Web3 payments look more like a back-office upgrade than a front-end revolution
One of the clearest conclusions from the last six months, in her view, is that Web3 payments will not first scale through end users. They will not break out because users suddenly start adopting wallets. They are more likely to scale because companies begin upgrading treasury, reconciliation systems, cross-border settlement routes and capital pool management behind the scenes.
Put differently, the mainstream path may be Web2 at the front end and Web3 reconstruction at the back end. Liu described that as a hidden upgrade. That also means success depends more on system stability, compliance certainty and long-run operating capacity than on market education.
Geographic growth is uneven
She also argued that the next major expansion may not come from the most mature markets. Asia-Pacific is already relatively mature, while larger structural growth may appear in Latin America, Africa, the Middle East and South Asia, where payment systems are more fragmented, costs are higher, routes are more complex and users and merchants may be more willing to move.
The other side of those markets, however, is heavy localization, stronger regulatory differences and demanding operations. They do not reward cleverness alone, she wrote. They require long-term work on the ground.
When she put those opportunities together, one conclusion became unavoidable. Payments remains an attractive business, but the resource base it demands — long-term banking relationships, a mature and sustainable compliance system, risk controls that can survive stress tests and credibility accumulated through repeated engagement with regulators — sits outside her team’s present capability boundary.
That was not, she wrote again, a denial of the direction. It was respect for reality. The battlefield in payments still exists. It is simply not under her team’s feet.
What came after stepping away
When Liu finally decided not to keep building in Web3 payments, she said the moment did not feel like a dramatic ending. It felt more like an exploration reaching the point where it should stop. She did not leave the industry. She moved from trying to stand on the waterway and collect the flow to standing beside it and watching more carefully how money moves and where it ends up.
As she kept unpacking payment structures, one judgment became sharper: payments solve the problem of movement. They determine whether money can move and how fast. But long-term value is not determined by movement itself. It is determined by where money stays after it moves and how it is managed.
Looking back over roughly the past twenty years of Chinese fintech development, she wrote, the logic is clear. Payments are only the entrance. Balances are a transit station. What eventually creates scale and defensibility is the system for capital management and asset allocation that comes after. She cited Yu’e Bao, Tiantian Fund and Tianhong as examples, arguing that they did not become important because they did payments better. They stood after payments and reorganized capital flows that had already reached scale.
Payments are the doorway, she wrote, but not the destination. When she applies that structure to Web3, she sees a similar issue emerging. On-chain, a growing set of asset forms already exists that are not aggressive but are steady enough: lending, short-duration RWA, neutral strategies and portfolio-style products. In her comparison, they resemble on-chain money market funds, short-duration bond funds and steady allocation tools.
The problem, in her view, is not whether those assets exist. It is that most people do not know what kind of risk they are facing, and they do not have a clear entry point for understanding, comparing and judging those assets.
As more capital begins to move on-chain, she wrote, that problem will only become more obvious. At that point she realized that even without continuing in payments, she could remain part of the change in another way. Rather than competing for the waterway itself, she wants to explain the structure of the flow, lay out the boundaries and risks, and help people see which places are worth staying in and which require extra caution. She said that is the direction she and her team plan to continue exploring.
Liu closed by saying the article is not an attempt to deliver a final verdict on Web3 payments, nor to tell anyone to enter or exit. It is an effort to explain clearly why she chose not to continue, in the hope that others might at least take it as one reference point and perhaps avoid some detours.