Author: Gang Ge
After the previous article was published, many people sent me private messages, and the questions mainly fell into the following categories:
"This platform looks similar, is it reliable?"
"Wouldn't digital currency payments be less complicated?"
"Is it necessary for Airwallex to make its payment system so complex?"

Figure 1. Jack Zhang's original tweet.
This article not only explains why Airwallex chose the "asset-heavy" path, but also exposes a long-standing, unspoken problem in the global payments industry.
01. Homogeneous surface, differentiated underlying structure
When choosing a payment platform, corporate clients are often confused by one question: several payment companies they consult seem to have similar capabilities.
For example, hundreds of payment companies around the world use similar rhetoric to introduce their products: instant payment, global coverage, and service to modern enterprises. Even their functions and interfaces are becoming increasingly similar.
Everyone has global acquiring: by connecting to a packaged Visa/MasterCard channel, you can claim to "support 200+ countries and regions worldwide";
Everyone has a global account: by partnering with several banks, you can say "one account for global payments, covering 20+ mainstream currencies".

Figure 2. Surface homogeneity cannot conceal the huge differences in the underlying layers.
Users cannot see the real differences between platforms in the product introduction, which is why they repeatedly nitpick with payment institutions about costs, backgrounds, licenses, and risks.
02 Three Paths to Global Payments
Now that the front-end team has a clear understanding of the core capabilities, it is necessary to go back to the bottom layer and break down the common paths in this industry.
If you break down the mainstream players in the industry, there are roughly three paths.

Figure 3. Three diverging paths in the evolution of global payments
2.1 The first path: bypassing traditional links
The first option: Web3 digital currency payments
The common narrative along this path is the same: stablecoins, on-chain settlement, programmable payments, and peer-to-peer payments. Compared to traditional payments, these offer shorter paths, faster speeds, and lower costs; and they aim to penetrate the consumer retail market through a story of small, high-frequency transactions.
However, you'll find that very few players have emerged from this path of digital currency payments.

Figure 4. First Path: On-chain Settlement Bypassing Traditional Links
Technical aspects: Global payments arrive instantly; merchant settlement to card D1/D0 options available;
Service aspect: The payment cost is not high due to intense competition, and the mature local operation team provides services to customers to the last mile;
From a compliance perspective: various jurisdictions have consistently questioned its compliance, resulting in significant compliance friction.
At the product level: mainstream payment platforms also focus on stablecoin payments. As long as compliance policies are implemented, products can be integrated and replaced at any time.
This is also why many people who dreamed of starting a business with Web3 payments ultimately had to give up in disappointment.
2.2 The Second Path: Repackaging Traditional Infrastructure
This is the most common path in the industry: relying on partners and intermediaries to repackage the complex and outdated underlying architecture, and then using better product experiences and faster marketing to drive market expansion.
The advantages of this path are also obvious: it is quick to produce results, quick to expand business, and quick to cover and expand coverage, so it will naturally become the choice of most players.
The problem is that it focuses more on optimizing the front end than rewriting the underlying layer.

Figure 5. Second Path: Aggregator Gateway Packaged Infrastructure
In fact, Stripe, the company with the fastest market capitalization growth in the world, had its acquisition of Airwallex rejected in 2019, and was rumored to be acquiring PayPal in 2026 (which was also ultimately rejected).
This at least illustrates one point: even international payment giants that have grown rapidly by relying on "technology + asset-light connectivity" will eventually have to go back and learn about infrastructure.
Often, a road that appears to be lighter doesn't mean it doesn't require infrastructure; it simply means that infrastructure construction has been postponed .
2.3 The Third Path: Building Our Own Global Financial Infrastructure
This path is the most difficult because there are almost no shortcuts. It requires sustained high investment, a longer timeframe, and also means heavier responsibilities.

Figure 6. The Third Path: Building Global Infrastructure Independently
03 The platform's "heavy" nature is exchanged for the customer's "light" nature.
For example, a market that has already been successfully established may suddenly have its bank account frozen; a customer may have already paid for the goods, but the funds are stuck halfway by the agent and cannot be credited to the account for a long time; when regulatory rules are upgraded, additional materials, additional deposits, and additional procedures must be completed.
These problems may not occur every day, but just one occurrence is enough to disrupt a company's rhythm.

Figure 7: The platform's "heavy" approach replaces the customer's "light" approach.
04 What did the customer really get?

Figure 8. The real value that enterprise customers need
It is more stable because companies do not need to adapt to a new set of partnerships every time they enter a new market.
It saves more than just transaction fees; it eliminates a significant amount of redundant system construction costs, communication costs, and compliance costs.
This is because when the market environment changes and regulations tighten, customers rely not on a temporary, cobbled-together channel, but on a more complete, durable, and sustainable underlying capability.
This is why Airwallex's growth logic is more like compound interest than explosive growth.

Figure 9. Compound growth brought about by infrastructure investment
According to publicly available information, it took Airwallex nine years to achieve $500 million in annualized revenue (ARR), but only one year to go from $500 million to $1 billion. The initial "slowness" was not inefficiency, but rather the accumulation of underlying potential energy for the subsequent acceleration.
05. In conclusion
Returning to the initial question, why did Airwallex build its own global financial infrastructure?

Figure 10. Underlying capabilities are the real watershed.
For corporate clients, choosing a global payment platform is essentially choosing a long-term partner who can help them manage complexity and provide a more stable foundation for their business.
[Reference Materials]
[1] Airwallex's official Weibo post: The difficult road is the only way out.
https://www.airwallex.com/cn/blog/the-path-of-max-resistance-the-spectrum-of-global-payments-infrastructure
[2] Airwallex official Weibo post: The last mile of global payments
https://www.airwallex.com/cn/blog/the-last-mile-of-global-payments
[3] Sina: Reshaping the Future of the Financial Industry
https://finance.sina.com.cn/cj/2025-10-11/doc-inftnpwr8889861.shtml




