Written by Dmitriy Berenzon

When Satoshi unveiled Bitcoin in 2009, he had the vision of using cryptonetworks for payments that could flow as freely as information across the internet. While this was directionally correct, the technology, economic model, and ecosystem were simply not well-suited to commercialize the use case. 

Fast forward to 2025, and we're witnessing the convergence of several important innovations and developments that make that vision inevitable: stablecoins have achieved widespread adoption by both consumers and businesses, market makers and OTC desks are now comfortably holding stablecoins on balance sheet, DeFi applications have created robust onchain financial infrastructure, a plethora of on/off-ramps exist around the world, blockspace is faster and cheaper, embedded wallets have simplified user experience, and clearer regulatory frameworks have reduced uncertainty.

An extraordinary opportunity exists today to build a new generation of payment companies that harness the power of "cryptorails" to achieve dramatically better unit economics than traditional systems burdened by multiple rent-seeking intermediaries and antiquated infrastructure. These cryptorails are forming the backbone of a parallel financial system that operates in real-time, 24/7, and is inherently global.

In this piece, I will:

• Explain the key components of the traditional financial system.

• Provide an overview of the major use cases for cryptorails.

• Discuss challenges preventing continued adoption.

• Share predictions on how the market will look in five years.

To further motivate this piece, it is important to note that there are many more companies operating here than you might think—about 320 as of this writing:

Existing Rails

In order to appreciate the significance of cryptorails, it is critical to first understand the key concepts of existing payment rails and the complex market structure and system architecture in which they operate. If you are familiar with this already then feel free to skip this section.

Card Networks

While the topology of card networks is complex, the key parties in a card transaction have stayed the same for the past 70 years. At its heart, a card payment involves four main players: 

1. Merchant

2. Cardholder

3. Issuing bank

4. Acquiring bank

The first two are straightforward but the last two are worth explaining.

The Issuing Bank or Issuer provides the credit or debit card to the customer and authorizes transactions. When a transaction request is made, the issuing bank decides whether to approve it by checking the cardholder’s account balance, available credit, and other factors. Credit cards essentially lend the issuer’s funds, while debit cards transfer money directly from your account.

If a merchant wants to accept card payments, they need an acquirer (which could be a bank, processor, gateway, or independent sales organization) that is a licensed member of the card networks. The term "acquirer" comes from their role of receiving money on behalf of merchants and ensuring that those funds reach the merchant's account.

The card networks themselves provide the rails and rules for card payments to occur. They connect acquirers to issuing banks, provide a clearing house function, set the rules of engagement, and determine transaction fees. ISO 8583 remains the primary international standard that defines how card-payment messages (e.g. authorizations, settlements, chargebacks) are structured and exchanged between network participants. In the context of networks, issuers and acquirers are like their distributors—issuers are responsible for getting more cards into the hands of users, and the acquirers are responsible for getting as many card terminals and payment gateways in the hands of merchants so they can accept card payments.

Furthermore, there are two flavors of card networks: “open loop” and “closed loop.” An open loop network like Visa and Mastercard involves multiple parties: issuing banks, acquiring banks, and the credit card network itself. The card network facilitates communication and transaction routing but acts as more of a marketplace, relying on financial institutions to issue cards and manage customer accounts. Only banks are allowed to issue cards for open loop networks. Each debit or credit card has a Bank Identification Number (BIN), which is offered by Visa to a bank, and non-bank entities like PayFacs need a “BIN sponsor” to be able to issue cards or process transactions. 

In contrast, a closed loop network like American Express is self-contained, with a single company handling all aspects of the transaction process—they usually issue their own cards, are their own bank, and provide their own merchant acquiring services. The general tradeoff is that closed loop systems offer more control and better margins at the cost of more limited merchant acceptance. Conversely, open loop systems offer wider adoption at the cost of control and revenue sharing for the involved parties.

The economics of payments are complex and there are several layers of fees in the network. Interchange is the portion of payment fees collected by the issuing bank in exchange for giving access to that issuer’s customers. While it is technically the acquiring bank that pays the interchange fee directly, the cost is usually passed along to the merchant. The card network usually sets the interchange fee and it usually makes up the bulk of the overall cost of a payment. These fees vary widely in different geographies and transaction types. For example, in the US, consumer credit card fees can range from ~1.2% to ~3%, while in the EU they are capped at 0.3%. In addition, scheme fees, also determined by the card network, are applied to transactions to compensate the networks for their role in connecting acquirers to issuing banks and acting as a “switch” to make sure the transactions and funds go to the correct parties. There are also settlement fees that go to the acquirer and are typically a percentage of the settled amount or volume of the transaction. 

While these are the most important parties along the value chain, the reality is that today’s market structure is much more complex in practice:

While I won’t go into all of them, there are a few important actors to call out:

A payment gateway encrypts and transmits payment information, connects with payment processors and acquirers for authorization, and communicates transaction approvals or declines to businesses in real-time.

A payment processor processes payments on behalf of the acquiring bank. It forwards transaction details from the gateway to the acquiring bank, which then communicates with the issuing bank via the card network for authorization. The processor receives the authorization response and sends it back to the gateway to complete the transaction. It also handles settlement, the process through which funds actually land in the merchant’s bank account. Typically, businesses send batches of authorized transactions to the processor, who submits them to the acquiring bank to initiate fund transfers from the issuing bank to the merchant’s account.

A payment facilitator (PayFac) or payment service provider (PSP), pioneered by PayPal and Square around 2010, is like a mini-payment processor that sits in between merchants and acquiring banks. It effectively acts as an aggregator by bundling many smaller merchants into their system to achieve economies of scale and streamline operations by managing the flow of funds, processing transactions, and ensuring payouts. PayFacs hold a direct Merchant ID with the card network and take on the responsibilities of onboarding, compliance (e.g. AML laws), and underwriting on behalf of the merchants they work with.

An orchestration platform is a middleware technology layer that streamlines and optimizes merchants' payment processes. It connects to multiple processors, gateways, and acquirers through a single API to improve transaction success rates, reduce costs, and enhance performance by routing payments based on factors like location or fees.

ACH

The Automated Clearing House (ACH) is one of the largest payment networks in the U.S. and is effectively owned by the banks that use it. It was originally formed in the 1970s but really took off when the U.S. government started using it to send Social Security payments, which encouraged banks across the country to join the network. Today, it is heavily used for payroll processing, bill payments, and B2B transactions.

ACH transactions have two main flavors: "push" payments (where you send money) and "pull" payments (where someone takes money with your permission). When you get your paycheck through direct deposit or pay a bill online using your bank account, you're using the ACH network. The process involves several players: the company or person starting the payment (the originator), their bank (ODFI), the receiving bank (RDFI), and operators who act like traffic controllers for all these transactions. In the ACH process, the originator submits a transaction to the ODFI, which then sends the transaction to an ACH operator, which then switches the transaction to the RDFI. At the end of each day, the operators calculate net settlement totals for their member banks (and the Federal Reserve manages the actual settlement).

One of the most important things to understand about ACH is how it handles risk. When a company initiates an ACH payment, its bank (the ODFI) takes responsibility for making sure everything is legitimate. This is particularly crucial for pull payments—imagine if someone used your bank account information without permission. To protect against this, regulations allow disputes up to 60 days after statement receipt, and companies like PayPal developed clever verification methods, such as making tiny test deposits to confirm account ownership.

The ACH system has tried to keep up with modern needs. In 2015, they introduced "Same Day ACH," which allows faster processing of payments. That said, it still relies on batch processing rather than real-time transfers and comes with limitations. For instance, you can't send more than $25,000 in a single transaction and it doesn't work well for international payments.

Wire

Wire transfers represent the backbone of high-value payment processing, with Fedwire and CHIPS as the two main systems in the U.S. These systems handle time-critical, guaranteed payments that need immediate settlement, such as securities transactions, major business deals, and real estate purchases. Once executed, wire transfers are generally irrevocable and cannot be canceled or reversed without the recipient's agreement. Unlike regular payment networks that process transactions in batches, modern wire transfer systems use real-time gross settlement (RTGS), meaning each transaction is settled individually as it occurs. This is an important property because the systems process hundreds of billions of dollars each day and the risk of an intraday bank failure using traditional net settlement would be too great.

Fedwire is an RTGS transfer system that allows participating financial institutions to send and receive same-day fund transfers. When an enterprise initiates a wire transfer, their bank verifies the request, debits the account, and sends a message to Fedwire. The Federal Reserve Bank then instantly debits the sending bank's account and credits the receiving bank's account, with the receiving bank subsequently crediting the final recipient. The system operates weekdays from 9 p.m. the prior calendar day to 7 p.m. Eastern Time and is closed during weekends and federal holidays.

CHIPS, owned by large U.S. banks through The Clearing House, serves as a private sector alternative but operates on a smaller scale, serving only a select group of major banks. Unlike Fedwire's RTGS approach, CHIPS is a netting engine, which means the system allows multiple payments between the same parties to be aggregated. For example, if Alice wants to send $10M to Bob and Bob wants to send $2M to Alice, CHIPs would consolidate these into a single payment of $8M from Bob to Alice. While this means that CHIPS payments take longer than real-time transactions, most payments still settle intraday.

Complementing these systems is SWIFT, which isn't actually a payment system but rather a global messaging network for financial institutions. It is a member-owned cooperative whose shareholders represent over 11,000 member organizations. SWIFT enables banks and securities firms worldwide to exchange secure, structured messages, many of which initiate payment transactions across various networks. According to Statrys, a SWIFT transfer takes about 18 hours to complete.

In the general flow, the sender of funds instructs their bank to send a wire transfer to a receiver. The value chain below is the simple case where both banks belong to the same wire transfer network.

In the more complex case, particularly with cross-border payments, the transaction needs to be executed through correspondent banking relationships, usually using SWIFT to coordinate the payments.

Use Cases

Now that we have a foundational understanding of traditional rails, we can focus on where cryptorails shine.

Cryptorails are most powerful in scenarios where traditional dollar access is limited but dollar demand is high. Think of places where people want USD for wealth preservation or as a bank alternative but can't easily get traditional USD bank accounts. These are generally countries experiencing economic instability, high inflation, currency controls, or underdeveloped banking systems, such as Argentina, Venezuela, Nigeria, Turkey, and Ukraine. In addition, one can argue that USD is a superior store of value relative to most other currencies and would be generally preferred by both consumers and businesses given the ability to easily use it as a medium of exchange or swap into local fiat at the point of sale.

The benefit of cryptorails is also greatest in scenarios where payments are global because cryptonetworks know no borders. They piggyback off of the existing internet connectivity, giving them global coverage out of the box. According to the World Bank, there are currently 92 RTGS systems operating around the world, each of which is generally owned by their respective central banks. While they are ideal for sending domestic payments within those countries, the issue is that they don’t “talk to each other”. Cryptorails can serve as the glue between these different systems as well as an extension into countries without them.

Cryptorails are also most useful for payments that have a degree of urgency or generally high time preference. This includes cross-border supplier payments and foreign aid disbursements. They are also helpful in corridors where the correspondent banking network is particularly inefficient. For example, despite geographical proximity, it is actually more difficult to send money from Mexico to the U.S. than from Hong Kong to the U.S. Even in developed corridors like the U.S. to Europe, payments can often go through four or more correspondent banks.

On the other hand, cryptorails are less compelling for domestic transactions within developed nations, especially where credit card adoption is high or where real-time payment systems already exist. For instance, intra-European payments work smoothly through SEPA, and the Euro's stability removes the need for dollar-denominated alternatives.

Merchant Acceptance

Merchant acceptance can be categorized into two distinct use cases: front-end integration and back-end integration. In the front-end approach, merchants can directly accept crypto as a form of payment from customers. While this is one of the oldest use cases, it has historically not seen much volume because few people held crypto, even fewer wanted to spend it, and limited useful options existed for those who did. The market is different today as more people hold cryptoassets, including stablecoins, and more merchants are accepting them as a payment option because it enables them to access new customer segments and ultimately sell more goods and services. 

From a geographical perspective, much of the volume is coming from businesses selling to consumers in countries that were early adopters of crypto, which are often emerging markets like China, Vietnam, and India. On the merchant side, much of the demand has been from online gambling and retail stock brokerages that want access to users in emerging markets, web2 and web3 marketplaces like watch vendors and content creators, and real-money gaming like fantasy sports and sweepstakes.

Here is what the “front-end” merchant acceptance flow generally looks like:

1. PSP creates a wallet for the merchant, often after KYC/KYB.

2. User sends crypto to PSP.

3. PSP off-ramps crypto into fiat via liquidity providers or stablecoin issuer and sends funds to the merchant’s local bank account, potentially using other licensed partners.

The main challenge preventing this use case from continuing to gain adoption is psychological because crypto doesn’t seem “real” to many people. There are two large user personas to address: one is completely detached from its value and wants to keep everything as magic internet money, and the other is pragmatic and off-ramps directly to their bank.

In addition, consumer adoption has been more difficult in the U.S. because credit card rewards are effectively paying consumers 1-5% on purchases. There have been attempts to sway merchants to promote crypto payments directly to consumers as an alternative payment method to cards altogether, however, they have not been successful to date. While lower interchange is a good pitch to merchants, it is not a problem for consumers. The Merchant Customer Exchange launched in 2012 and failed in 2016 for exactly this reason—they could not kick-start the consumer side of the adoption flywheel. In other words, getting users to switch from paying with credit cards to cryptoassets is very difficult for merchants to incentivize directly because payments are already "free" for consumers, so the value proposition should be solved at the consumer level first.

In the back-end approach, cryptorails can offer faster settlement times and access to funds for merchants. Settlement can take 2-3 days for Visa and Mastercard, 5 days for American Express, and even longer internationally, about 30 days in Brazil for example. In some use cases, such as marketplaces like Uber, the merchant might need to pre-fund a bank account to disburse payments prior to settlement. Instead, one can effectively on-ramp via a user's credit card, transfer funds onchain, and off-ramp directly to the merchant’s bank account in their local currency. In addition to the working capital improvement from this flow due to having less capital tied up in transit, merchants can further improve their treasury management by freely and instantly swapping between digital dollars and yield-bearing assets, such as tokenized U.S. Treasuries.

More specifically, here is what the “back-end” merchant acceptance flow might look like:

1. Customer enters their credit card information to complete a transaction.

2. PSP creates a wallet for the customer and user funds that wallet with an on-ramp that accepts traditional payment methods.

3. The credit card transaction buys USDC, which is then sent to the merchant’s wallet from the customer’s wallet.

4. PSP can optionally off-ramp to merchant’s bank account via local rails T+0 (i.e. same day).

5. PSP receives funds from acquiring bank usually T+1 or T+2 (i.e. in 1-2 days).

Debit Cards

The ability to link debit cards directly to non-custodial smart contract wallets has created a surprisingly powerful bridge between blockspace and meatspace, driving organic adoption across diverse user personas. In emerging markets, these cards are becoming primary spending tools, increasingly replacing traditional banks. Interestingly, even in countries with stable currencies, consumers are leveraging these cards to gradually build USD savings while avoiding foreign exchange (FX) fees for purchases. High-net-worth individuals are also increasingly using these crypto-linked debit cards as an efficient tool for spending their USDC worldwide.

The traction with debit over credit cards comes from two factors: debit cards face fewer regulatory restrictions (for example, MCC 6051 is outright declined in Pakistan and Bangladesh, which have strict capital controls), and they present lower fraud risk since chargebacks for already-settled crypto transactions create significant liability issues for credit cards.

Over the longer-term, cards tied to crypto wallets that are used for mobile payments might actually be the best way to combat fraud because of the biometric verification on your phone: scan your face, spend stables, and top up from your bank account to wallet.

Remittance

Remittance is the movement of funds from the country of work back to a home country for people who moved internationally to find work and want to send money back to their families. According to the World Bank, remittance volume in 2023 totaled about $656 billion, equivalent to Belgium's GDP. 

The traditional remittance system has significant costs that result in fewer dollars in the pocket of the recipient. On average, sending money across borders costs 6.4% of the transfer amount but these fees can vary dramatically—from 2.2% for transfers from Malaysia to India (and even lower for high-volume corridors like the U.S. to India) to an astounding 47.6% from Turkey to Bulgaria. Banks tend to be the most expensive, charging around 12%, while money transfer operators (MTOs) like MoneyGram average 5.5%.

Cryptorails can offer a faster and cheaper way to send money abroad. The traction of companies using cryptorails largely follows the broader remittance market sizes, with the highest-volume corridors being from the U.S. to LatAm (specifically Mexico, Argentina, and Brazil), U.S. to India, and U.S. to Philippines. An important enabler of this traction has been non-custodial embedded wallets like Privy, which offer a web2-grade UX for users.

The flow for a remittance payment using cryptorails might look like this:

1. The sender on-ramps to the PSP via a bank account, debit card, credit card, or directly to an onchain address; if the sender does not have a wallet, one is created for them.

2. The PSP converts the USDT/USDC to the recipient’s local currency, either directly or using a market maker or OTC partner.

3. The PSP disburses the fiat to the recipient’s bank account, either directly from their omnibus bank account or via a local payment gateway; alternatively, the PSP can also first generate a non-custodial wallet for the recipient to claim the funds, giving them the option of keeping it onchain.

4. In many cases, the recipient needs to first complete KYC to receive the funds.

That said, the go-to-market for crypto remittance projects is tough. One issue is that you generally need to incentivize people to shift from MTOs, which can be expensive. Another issue is that transfers are free on most web2 payment apps already, so local transfers alone are not compelling enough to overcome the network effects of existing applications. Lastly, while the onchain transfer component works well, you still need to interact with TradFi at the “edges”, so you might still end up with the same, if not worse, problems due to the off-ramping cost and friction. In particular, payment gateways that convert into local fiat and payout in bespoke ways like mobile phones or kiosks will take the most margin.

B2B Payments

Cross-border (XB) B2B payments are one of the most promising applications for cryptorails because the traditional system suffers from significant inefficiencies. Payments going through the correspondent banking system can take weeks to settle, with some extreme cases taking even longer—one founder said it took them 2.5 months to send a supplier payment from Africa to Asia. As another example, a cross-border payment from Ghana to Nigeria, two bordering countries, can take weeks and cost up to 10% transfer fees.

In addition, cross-border settlement is slow and expensive for PSPs. For a company that does payouts like Stripe, it can take them up to a week to pay a merchant internationally and they have to lock up capital to cover the risk of fraud and chargebacks. Cutting down the conversion cycle time would free up significant amounts of their working capital.

B2B XB payments have seen significant traction with cryptorails largely because merchants care more about fees than consumers. Shaving off 0.5-1% on transaction costs doesn’t sound like a lot, but it adds up when there’s a lot of volume, particularly for businesses operating on thin margins. Furthermore, speed matters. Getting a payment to settle in hours instead of days or weeks has a meaningful impact on a company’s working capital. In addition, businesses have more tolerance for worse UX and more complexity relative to consumers who expect a smooth experience out of the box.

Furthermore, the cross-border payments market is massive—estimates vary widely by source but according to McKinsey it was about $240 billion in revenue and $150 trillion in volume in 2022. That said, building a sustainable business can still be tough. While the “stablecoin sandwich”—converting local currency to stablecoins and back again—is definitely faster, it is also expensive because doing FX conversions on both sides erodes margins, often to the point of unsustainable unit economics. While some companies try to solve this by building internal market-making desks, this is very balance-sheet intensive and difficult to scale. In addition, the customer base is also relatively slower, concerned with regulation and risk, and generally requires a lot of education. That said, FX costs will likely rapidly decline over the next two years as stablecoin legislation opens up access for more businesses to hold and operate with digital dollars. Because more on/off-ramps and token issuers will have direct banking relationships, they will effectively be able to offer wholesale FX rates at internet scale.

XB Supplier Payments

With B2B payments, much of the cross-border volume has been around supplier payments for imports, where usually the buyer is in the U.S., LatAm, or Europe, and the supplier is in Africa or Asia. These corridors are particularly painful because the local rails in those countries are underdeveloped and difficult to access for companies because they cannot find local banking partners. There are also country-specific pain points that cryptorails can help alleviate. For example, in Brazil you cannot pay millions of dollars out using traditional rails, which makes it difficult for businesses conducting international payments. Some well-known companies, like SpaceX, are already using cryptorails for this use case.

XB Receivables

Businesses that have customers around the world often struggle to collect funds in a timely and efficient manner. They often work with multiple PSPs to collect funds for them locally but need a way to quickly receive them, which can take multiple days or even weeks depending on the country. Cryptorails are faster than a SWIFT transfer and can compress that time down to T+0.

Here is what an example payment flow might look like for a business in Brazil buying goods from a business in Germany:

1. Buyer sends Real to PSP via PIX.

2. PSP converts Real to USD, then on-ramps into USDC.

3. PSP sends USDC to the wallet of the seller.

4. If the seller wants local fiat, PSP sends USDC to a market maker or trading desk to convert into local currency.

5. PSP can send funds to seller via local rails if they have a license/bank account, or use a local partner if not.

Treasury Operations

Companies can also use cryptorails to improve their treasury operations and accelerate global expansion. They can hold USD balances and use local on/off-ramps to reduce FX exposure and enter new markets more quickly, even where local banking providers are reluctant to support them. They can also use cryptorails as an internal means of reorganizing and repatriating funds between the countries in which they operate.

Foreign Aid Disbursement

Another general use case we are seeing for B2B is time-critical payments, for which these cryptorails can be used to reach the recipient faster. One example is foreign aid payments—allowing NGOs to use cryptorails to send money to local off-ramp agents who can individually disburse payments to qualified individuals. This can be particularly impactful in economies with very poor local financial systems and/or governments. For example, countries like South Sudan have a collapsing central bank and local payments there can take over a month. But as long as there’s access to mobile phones and an internet connection, there is a way to get digital money into the country and individuals can exchange that digital money for fiat and vice versa.

The payment flow for this use case might look like this:

1. NGO sends funds to PSP.

2. PSP sends a bank transfer to OTC partner.

3. OTC partner on-ramps the fiat into USDC and sends it to a local partner's wallet.

4. Local partner off-ramps USDC via peer-to-peer (P2P) traders.

Payroll

From a consumer perspective, one of the most promising early adopters has been freelancers and contractors, particularly in emerging markets. The value proposition for these users is that more money ends up in their pockets rather than going to intermediaries and that the money can be in digital dollars. This use case also has cost benefits for the businesses on the other side who are sending large-scale payouts and is especially useful for crypto-native companies like exchanges that already hold most of their treasury in crypto.

The payment flow for a contractor payout generally looks like this:

1. Company does KYB/KYC with the PSP.

2. Company sends USD to PSP or USDC to a wallet address that’s tied to a contractor.

3. Contractor can decide if they want to keep it in crypto or withdraw to a bank account, and PSP often has some master service agreement with one or more off-ramp partners who hold the relevant licenses in their respective jurisdictions to do the local payout.

On/Off-ramps

On/off-ramps is a crowded market that has many corpses and zombies. While many early attempts failed to scale, the market has matured over the past several years, with many companies operating sustainably and offering access to local payment rails around the world. While on/off-ramps can be used as a standalone product (e.g. simply buying cryptoassets), they are arguably the most critical part of the payment flow for bundled services like payouts.

Building an on/off-ramp generally has three components: obtaining the necessary licenses (e.g. VASP, MTL, MSB), securing a local banking partner or PSP for access to local payment rails, and connecting with a market maker or OTC desk for liquidity. 

On-ramping was initially dominated by exchanges, but today, an increasing number of liquidity providers, from smaller FX and OTC desks to larger trading firms like Cumberland and FalconX, are offering ramping. These firms can often handle up to $100M/day in volume, which makes it unlikely that they will run out of liquidity for popular assets. Some teams might even prefer them because they can promise spreads, which helps control margins.

The non-U.S. leg of an off-ramp is usually much more difficult than the U.S. leg of an on-ramp due to licensing, liquidity, and orchestration complexity. This is particularly the case in LatAm and Africa, where there are dozens of currencies and payment methods. For example, you can use PDAX in the Philippines because it is the largest crypto exchange there, but in Kenya you need to use several local partners like Clixpesa, Fronbank, and Pritium depending on the payment method. 

P2P ramps rely on a network of "agents"—local individuals, mobile money providers, and small businesses like supermarkets and pharmacies—who provide both fiat and stablecoin liquidity. These agents, particularly prevalent in Africa where many already operate mobile money stalls for services like MPesa, are motivated primarily by economic incentives—they earn through transaction fees and FX spreads. In fact, for individuals in high-inflation economies like Venezuela and Nigeria, becoming an agent can be more lucrative than traditional service jobs like taxi drivers or food delivery. They can also just work from home using their mobile phone, and usually just need a bank account and mobile money to get started. What makes this system particularly powerful is its ability to support dozens of local payment methods with zero formal licenses or integrations since transfers happen between individual bank accounts. 

Notably, the FX rate with P2P ramps can often be significantly more competitive. For example, the Bank of Khartoum in Sudan often charges up to 25% for FX fees while local crypto P2P off-ramps are offering 8-9%, which is effectively the market rate instead of the bank-enforced rate. Similarly, P2P ramps are able to offer FX rates that are about 7% cheaper than bank rates in both Ghana and Venezuela. Generally, the spread is smaller in countries where there is greater availability of USD. Furthermore, the best markets for P2P ramps are those with high inflation, high smartphone adoption, poor property rights, and unclear regulatory guidelines since financial institutions won’t touch crypto, which creates an environment for self-custody and P2P to thrive.

Here is how a payment flow for a P2P on-ramp might look:

1. User either selects or is automatically assigned a counterparty, or an “agent”, who already has USDT, which is often escrowed by the P2P platform.

2. User sends fiat to an agent via local rails.

3. Agent confirms receipt and USDT is sent to the user.

From a market structure perspective, most on/off-ramps are commoditized and there is little customer loyalty, as they will usually choose the cheapest option. To remain competitive, local ramps will likely need to expand coverage, optimize for the most popular corridors, and find the best local partners. In the long-term, we will likely see consolidation into a few on/off-ramps in each country that each have comprehensive licensing, support for all local payment methods, and offer the most liquidity. In the mid-term, aggregators will be particularly useful because local providers are often faster and cheaper, and combining options will often offer the best pricing and completion rates for consumers. They might also suffer the least commoditization if they can efficiently optimize and route payments across hundreds of partners and routes. This also applies to orchestration platforms that can include compliance, PSP selection, bank partner selection, and value-added services such as card issuance.

From a consumer perspective, the good news is that fees will likely trend to zero. We already see this today with Coinbase, where it costs $0 to instantly move from USD to USDC. In the long-run, most stablecoin issuers will likely enable this for large wallets and FinTechs, further compressing ramp fees.

Licensing

Licensing is a painful but necessary step for scaling the adoption of cryptorails. There are two approaches for startups: partner with an already-licensed entity or obtain licenses independently. Working with a licensed partner allows a startup to bypass the substantial costs and lengthy timelines associated with obtaining a license themselves, but at the cost of tighter margins due to a substantial portion of the revenue going to the licensed partner. Alternatively, a startup can choose to invest upfront—potentially hundreds of thousands to millions of dollars—to acquire a license independently. While this path often takes months and even years (one project said it took them 2 years), it enables the startup to offer a more comprehensive product directly to users. 

While there are established playbooks for obtaining licenses across many jurisdictions, achieving global licensing coverage is exceptionally challenging, if not impossible, as each region has its own unique regulations around money transmission and you would need over 100 licenses for global coverage. For example, just in the U.S., a project would need a Money Transmitter License (MTL) for each state, a BitLicense for New York, and a Money Services Business (MSB) Registration with the Financial Crimes Enforcement Network. Just getting MTLs for all states can cost from $500K to $2M and can take up to a year. The requirements are equally as dizzying when looking overseas—a good resource can be found here. Importantly, startups that are non-custodial and are not touching the flow of funds can generally bypass immediate licensing requirements and get to market faster.

Challenges

Payments adoption is usually difficult because they’re a chicken-and-egg problem. You need to either get widespread consumer adoption of a payment method, which will force merchants to accept it, or have merchants use one specific payment method, which will force consumers to adopt it. For example, credit cards were niche in LatAm until Uber started to take off in 2012; everyone then wanted a credit card because it would allow them to use Uber which was much safer and (initially) cheaper than cabs. This allowed other on-demand apps like Rappi to take off because there were now people with smartphones and credit cards. This became a virtuous cycle where more people wanted a credit card because there were more cool applications that required one for payment. 

This also applies to mainstream consumer adoption of cryptorails. We have yet to see a use case where it is particularly advantageous or completely necessary to pay with stablecoins, although debit cards and remittance applications are getting us closer to that moment. A P2P app also has a chance if it unlocks a net new behavior online—micropayments and creator payments seem like exciting candidates. This is broadly true of consumer applications in general, where adoption won’t happen without step function improvements over the status quo.

There are also several problems that continue to exist for on/off-ramping:

• High failure rates: You understand this frustration if you have tried using a credit card to on-ramp.

• UX friction: While early adopters were OK with the pain of onboarding with exchanges to access assets, the early majority will likely be onboarded natively within specific applications. To support this, we need smooth in-app ramping, ideally via Apple Pay.

• High fees: On-ramping is still very expensive—it can still cost 5-10%, depending on the provider and region.

• Inconsistent quality: There is still too much variance in reliability and compliance, especially with non-USD currencies.

An under-discussed issue is privacy. While it is not currently a serious concern for either individuals or companies, it will become one once cryptorails are adopted as the primary mechanism for commerce. There will be serious negative consequences when malicious actors start to monitor payment activity for individuals, companies, and governments via their public keys. One way to address this in the short term is to do “privacy by obscurity” and spin up new wallets each time funds need to be sent or received onchain.

In addition, establishing banking relationships is often the most difficult part because it’s another chicken and egg problem. Banking partners will take you if they get transaction volumes and will make money, but you need banks to get those volumes in the first place. In addition, there currently are only 4-6 small U.S. banks that support crypto payments companies today, and several are hitting their internal compliance limits. Part of this is because crypto payments today are still categorized as “high-risk activities” similar to cannabis, adult media, and online gambling.

Contributing to this issue is the fact that compliance is still not at par with traditional payment companies. This includes AML/KYC & Travel Rule compliance, OFAC screening, cybersecurity policies, and consumer protection policies. Even more challenging is directly baking compliance into cryptorails rather than relying on out-of-band solutions and companies. Lightspark's Universal Money Addresses offers one creative solution to this challenge by facilitating the exchange of compliance data between participating institutions.

Future Outlook

On the consumer side, we are currently at the point where certain demographics are hitting mainstream adoption, particularly with freelancers, contractors, and remote workers. We are also getting closer to mainstream in emerging economics with USD demand by piggybacking off of the card networks and offering consumers USD exposure along with the ability to spend it for everyday use. In other words, debit cards and embedded wallets have become the “bridge” that is bringing crypto offchain in a form factor that is intuitive for mainstream consumers. On the business side, we are firmly at the start of mainstream adoption. Companies are using stablecoins at scale and this will significantly increase over the decade.

With all of this in mind, here are my 20 predictions on how the industry might look in 5 years:

1. $200-$500 billion in payments volume annually via cryptorails, primarily driven by B2B payments.

2. >30 neobanks around the world launched natively on top of cryptorails.

3. Dozens of acquisitions of crypto-native companies as FinTechs race to remain relevant.

4. Several crypto companies, likely stablecoin issuers, will acquire struggling FinTechs and banks who run out of runway due to high CAC and operational costs.

5. ~3 cryptonetworks (both L1s and L2s) emerge and hit scale with an architecture that is specifically designed for payments. Such a network will resemble Ripple in spirit but will have a tech stack, economic model, and go-to-market that makes sense.

6. 80% of online merchants will be accepting crypto as a means of payment, either through existing PSPs expanding their offering or crypto-native payment processors offering a better experience for them.

7. Card networks will expand to cover ~240 countries and territories (up from ~210 today) using stablecoins as the last-mile solution.

8. >15 remittance corridors around the world will have the majority of their volume flow via cryptorails.

9. Onchain privacy primitives will finally be adopted, driven by businesses and countries using cryptorails rather than consumers.

10. 10% of all foreign aid disbursements will be sent via cryptorails.

11. The on/off-ramp market structure will ossify, with 2-3 providers per country gaining the majority of volumes and partnerships.

12. There will be as many P2P on/off-ramp liquidity providers as food delivery workers in the countries where they operate. As volumes increase, agents will become economically sustainable jobs and continue to be at least 5-10% cheaper than bank-quoted FX rates.

13. >10 million people who are remote workers, freelancers, and contracts will receive payment (either directly in stablecoins or local currency) for their services via cryptorails.

14. 99% of AI agent commerce (including agent-to-agent, agent-to-human, and human-to-agent) will happen onchain via cryptorails.

15. >25 reputable partner banks in the US will support companies operating on top of cryptorails, removing the bottlenecks that operation chokepoint exacerbated.

16. Financial institutions will experiment with issuing their own stablecoins to facilitate global real-time settlement.

17. Standalone “crypto Venmo” applications still do not take off as the user persona remains too niche, but large messaging platforms like Telegram will integrate cryptorails that will start being used for P2P payments and remittance.

18. Lending and credit companies will start collecting and disbursing payments via cryptorails to improve their working capital due to less funds tied up in transit.

19. Several non-USD stablecoins will begin being tokenized at scale, giving rise to onchain FX markets.

20. CBDCs remain an experiment and do not hit commercial scale because of government bureaucracy.

Conclusion

As Patrick Collison alluded to, cryptorails are superconductors for payments. They form the substrate of a parallel financial system that offers faster settlement times, reduced fees, and the ability to operate seamlessly across borders. It took a decade for the idea to mature, but today we see hundreds of companies working to make it a reality. Over the next decade, we will see cryptorails at the heart of financial innovation, driving economic growth across the globe. 

If you are building something using cryptorails, please reach out. You can find me on Twitter/X and Farcaster.

Many thanks to Jeremy Allaire, Sam Broner, Christian Catalini, Katie Chiou, Wyatt Lonergan, and DC Posch for conversations and feedback on this piece.

Disclaimer:

This post is for general information purposes only. It does not constitute investment advice or a recommendation or solicitation to buy or sell any investment and should not be used in the evaluation of the merits of making any investment decision. It should not be relied upon for accounting, legal or tax advice or investment recommendations. You should consult your own advisers as to legal, business, tax, and other related matters concerning any investment or legal matters. Certain information contained in here has been obtained from third-party sources, including from portfolio companies of funds managed by Archetype. This post reflects the current opinions of the authors and is not made on behalf of Archetype or its affiliates and does not necessarily reflect the opinions of Archetype, its affiliates or individuals associated with Archetype. The opinions reflected herein are subject to change without being updated.