How the CAIP-358 Standard Enables Direct Self-Custodial Wallet Payments in Retail

For years, the primary criticism of cryptocurrency payments at point of sale has been unacceptable user experience. The process required four to six separate user interactions: token selection, blockchain network selection, waiting for address generation, manual amount entry, transaction confirmation. For someone standing at a checkout counter, this scenario excluded such payment methods from everyday use.

In February 2025, customers at Metro Department Store in Singapore began paying for purchases by scanning a QR code and clicking one confirmation button in their self-custodial wallet. Transaction time: ten to fifteen seconds. User experience comparable to Alipay or Apple Pay. This is the first production deployment of the CAIP-358 standard (Universal Payment Request Method), which fundamentally changes how wallets communicate with merchant systems.

I'll show you how this standard works technically, why it solves problems of previous implementations, and what it means for retail business owners and users actively utilizing self-custodial wallets in the Web3 ecosystem.

Why Previous Solutions Failed in Retail

Existing blockchain payment standards like EIP-681 for Ethereum or BIP-21 for Bitcoin had two fundamental limitations. First, each was specific to a single blockchain ecosystem. A merchant wanting to accept payments on Ethereum, Solana, and Polygon had to implement three separate solutions. Second, different payment providers developed proprietary, incompatible protocols for wallet-to-point-of-sale communication.

A typical pre-CAIP-358 payment flow looked like this: user informed merchant which cryptocurrency to use, then which blockchain network, system generated payment address (sometimes taking five to fifteen seconds), user opened wallet application, scanned QR code, approved connection with merchant application, manually entered amount, confirmed transaction, and waited for on-chain confirmation. Each of these steps represented a potential point of failure or process interruption.

According to WalletConnect's "The State of Onchain Payments 2025" report, only ten percent of cryptocurrency users indicated payments as their preferred use case. The main obstacle cited was inefficiency – too many steps, too long execution time, too high probability of error.

Technical Architecture of CAIP-358

Luka Isailovic and Derek Rein developed a standard that transfers all selection complexity from user interaction level to automatic wallet processing. The key change is that the merchant sends the wallet an array of all accepted payment options in a single request.

This request structure contains an order identifier, an array of possible payment methods, and request expiration time. Each payment method specifies recipient address, asset identifier according to CAIP standards, and amount. For example, a merchant can specify in one request that they accept USDC on Ethereum, USDC on Polygon, and USDT on Solana.

Upon receiving such a request, the wallet automatically filters options based on assets the user actually possesses. The selection algorithm considers available balance, current network fees, and user preferences. The user sees a single screen: "Pay 47.50 USDC to Metro Department Store" with a confirmation button. One interaction instead of six.

After confirmation, the wallet executes the transaction and returns to the merchant system a structure containing the blockchain transaction identifier and confirmed payment parameters. The standard defines an idempotency mechanism – if payment for a given order identifier has already been completed, the wallet returns the original result without executing a new transaction. This solves the connection loss problem during payment, where the merchant system can retry the request without risk of double payment.

A significant element is privacy by design. The user's wallet address is not transmitted to the merchant before the transaction. The wallet independently decides which address to use. This eliminates the possibility of merchants tracking users based on address as an identifier across different transactions.

First Production Implementation in Singapore

Singapore-based dtcpay, holding a Major Payment Institution license from the Monetary Authority of Singapore, launched this system at Metro Department Store in February 2025. This is a significant case for one reason: it demonstrates the standard works in actual retail conditions with real customers and real transactions. Not a pilot, not proof of concept, but production implementation in a major department store.

The integration process from the merchant's perspective does not pose a major technical challenge. The POS terminal generates a payment request according to CAIP-358 specification and displays a WalletConnect QR code. That's all. WalletConnect infrastructure handles the rest – encrypted wallet connection, message exchange, confirmation. You'll soon see similar implementations in other locations worldwide, as the barrier to entry for merchants is low.

According to Chainalysis, stablecoin transactions in Singapore reached nearly one billion dollars in value in Q2 2024. These are actual payments for goods and services, not trading volumes on exchanges. On the WalletConnect network, stablecoins constitute seventy-two percent of all payments, with USDC accounting for thirty-eight percent and USDT for thirty-four percent. These numbers demonstrate that demand for such solutions is real.

Implications for Self-Custodial Wallets

WalletConnect currently supports over seven hundred wallets, including applications with self-custodial wallet functionality such as mone.my. Each of these applications that implements the wallet_pay method according to CAIP-358 standard automatically gains the ability to make payments at points of sale cooperating with processors supporting this standard.

This represents a key change in business model. Previously, each wallet had to negotiate separate agreements with merchants or payment processors. Now a merchant integrating with a processor supporting CAIP-358 gains compatibility with all wallets implementing the standard. For a mone.my user or another self-custodial application, this means they can pay anywhere a merchant displays a WalletConnect QR code, without requiring prior configuration or registration with specific merchants.

The standard does not enforce a specific business model on the processor side. It can offer immediate conversion of received stablecoins to fiat currency, storage in stablecoins, or hybrid solutions. Transaction costs depend on the chosen blockchain layer. Transactions on Polygon generate network fees in the range of one to five cents, Ethereum mainnet from one to five dollars depending on network congestion, Solana below one cent.

For a user possessing a self-custodial wallet with WalletConnect functionality, the in-store payment process differs from card payment only in requiring opening the application and scanning the code. Execution time remains comparable: three to eight seconds for Polygon, twelve to fifteen seconds for Ethereum, one to two seconds for Solana. The difference lies in the settlement layer: card payment requires intermediation of acquirer, issuer, and card network with final settlement occurring one to three days later, while stablecoin payment executes immediate on-chain settlement directly to the merchant's address.

WalletConnect Network Scale and Shifting Payment Infrastructure Dynamics

WalletConnect is set to surpass four hundred billion dollars in annual Total Network Volume. This positions the network on a scale comparable to major global fintech players. For context: Square processes two hundred thirty-one billion dollars annually in gross payment volume, Checkout.com projects three hundred billion dollars, Shopify Payments handles two hundred ninety-two billion dollars in gross merchandise value, and Wise facilitates one hundred forty-five billion dollars in cross-border volume.

The network currently serves three hundred fifty million wallet-to-app connections across fifty million users, seventy thousand applications, and seven hundred wallets. Institutional participants include Fireblocks, Ledger, Robinhood, Blockchain.com, OKX Wallet, Binance Wallet, and Gemini Wallet. Applications such as Aave, Spark.fi, and Hyperliquid drive significant volume, while enterprises like Stripe, Coinbase Commerce, and Shopify utilize WalletConnect for onchain payment infrastructure.

This scale fundamentally changes the context for CAIP-358 retail deployment. When I write about stablecoin payments entering physical retail through this standard, I'm describing the opening of a new segment for infrastructure already processing hundreds of billions in annual volume. The retail payment capability represents logical expansion from DeFi protocols and dApp interactions into physical commerce.

The power dynamics in payment infrastructure are shifting. Traditional payment processors built networks over decades by establishing merchant relationships and issuing cards to consumers. WalletConnect built a network by becoming connectivity standard for blockchain applications and wallets. The four hundred billion dollar annual volume flows through this infrastructure without WalletConnect ever holding custody of assets or controlling private keys. Every connection operates with end-to-end encryption, fundamentally different architecture than traditional payment rails.

For retail business owners evaluating whether to integrate stablecoin payment capabilities, the relevant question is not whether blockchain-based payment infrastructure will scale to handle retail volumes. It already processes four hundred billion dollars annually. The question is whether your customer base includes users of the seven hundred wallets in this network, and whether the economics of immediate on-chain settlement versus traditional card processing justify integration costs.

The trajectory suggests movement from billions to trillions in annual volume. As institutional adoption accelerates and regulatory frameworks mature in jurisdictions beyond Singapore, retail represents significant addressable market expansion for this infrastructure. CAIP-358 provides the technical standard making this expansion operationally viable by solving the user experience problems that previously made blockchain payments impractical at physical points of sale.

Coexistence with Traditional Payment Infrastructure

The stablecoin payment system via self-custodial wallet does not replace card payments. It functions as a parallel path for users possessing such wallets and preferring this payment method. Mastercard in June 2025 announced the Web3 Card program enabling self-custodial wallet holders to issue payment cards, where conversion from cryptocurrency to fiat occurs at the moment of transaction authorization. Visa is developing similar solutions.

The Tempo blockchain, which I wrote about in a previous article, implements interbank settlements using blockchain as technical infrastructure. CAIP-358 and WalletConnect Pay represent a different path – the entire payment flow from user to merchant occurs directly on-chain, without intermediation of traditional payment institutions in the settlement layer.

These models can coexist in a single terminal. A standard terminal with NFC reader supports card and mobile payments, the same terminal displaying a WalletConnect QR code supports stablecoin payments. From my perspective as someone observing payment system transformation, this is not competition between models but different implementation paths for a fundamental change: transition from legacy payment rails to blockchain-based infrastructure for value transfers.

Limitations and Future Development Directions

The CAIP-358 standard remains in draft status and requires formal acceptance by the Chain Agnostic Standards Alliance. Production implementation precedes standard finalization, which is typical for standardization driven by real-world deployment.

The current specification does not cover recurring payment mechanisms, partial payments, or dispute resolution. It also does not define merchant reputation protocols or on-chain review systems. These functionalities will likely be subjects of future extensions.

Technical challenges concern scalability with mass adoption. Ethereum mainnet processes fifteen transactions per second, which represents a limitation for retail payment volumes. Layer 2 solutions offer higher throughput but require users to hold assets on specific networks. Cross-chain interoperability – where a wallet holding USDC on one chain automatically bridges to another required by the merchant – is not currently part of the standard and generates additional costs and latency.

The regulatory landscape remains geographically diverse. Singapore, through clear regulatory frameworks for payment institutions handling digital payment tokens, enables rapid commercialization. The European Union is implementing MiCA, which defines requirements for stablecoin issuers and payment service providers. The United States lacks coherent federal regulations, slowing deployment in the world's largest economy.

The CAIP-358 standard reduces cryptocurrency payment user experience in retail to a level comparable with card payments. I'm showing you this solution because the fundamental change consists of transferring complexity from the user interaction level to automatic wallet processing. The process reduces to a single interaction: scan code and confirm. For users actively utilizing self-custodial wallets, this eliminates the main barrier to payment adoption at physical points of sale. For retail business owners, this signals that blockchain payment infrastructure has reached scale – four hundred billion dollars in annual network volume – where retail integration becomes strategically relevant rather than experimental. The technical barrier to entry is low, the potential user base spans seven hundred compatible wallets, and the shift in payment infrastructure dynamics is already underway.