This is an excerpt from the 0xResearch newsletter. To read the full article, register.
One of the new features of Hyperliquid is building codes. These tokens act as part of the protocol-level parameter in the payload, allowing interfaces to add address builders for standard, onchain functionality. Builders can include interest of up to 100 basis points (1%) on real estate and 10 basis points (0.1%) on perps.
This integration allows front-ends to access equity capital without the technical difficulties of maintaining an order book or the financial constraints of bootstrapping liquidity. As shown below, the third-party front-end includes Hyperliquid perps and adds their top-paying shares, effectively creating a difference in the price of the corresponding shares.
As such, building codes have turned on the power distribution flywheel. About 40% of daily users are now shopping through other parts instead of the standard UI, a share that briefly passed 50% at the end of October. The top three builders alone, Based, Phantom, and pvp.trade, together took in more than $31 million.
From a market perspective, this pushes Hyperliquid away from a more integrated form of crypto exchange and closer to the integration of traditional assets. On a centralized exchange like Binance, a single team controls the entire volume of transactions, transactions, matching and storage.
The structure of Hyperliquid imitates the US equity market, where brokers (Robinhood, Schwab) have customer relations and distribute funds, while ordering brokers (Citadel Securities, Virtu) handle operations and returns. Basically, a stack is of two types:
- A broker-like distribution segment, where builders compete for order flow and differentiate on price and transit fees.
- The middle of the kill zone, where Hyperliquid focuses on liquids and controls the comparison and limits.
Although new to crypto perps, this link has already played out on Solana. Vendors such as Photon and Axiom dominated the user experience by focusing on consumers. Photon developed first by being the fastest sniper of the Solana memecoin, while Axiom eventually challenged it with a bunch of features and aggressive policies and refunds. These sites functioned effectively as builders: They sat on top of DEXs, building on their payments, and keeping money on hand. The Hyperliquid build code changes the color to the original.
However, Solana’s example also presents a danger. Stocks accounted for 77% of Solana’s DEX revenue last year, $633 million versus $188 million for DEXs, a 3.4x increase that shows that having a front-end is often more important than having a backend. In particular, is the front most important for Hyperliquid to deliver?
The relationship between frontends and backends is often not symbiotic. Frontends like Jupiter combine different backends (Meteora, Raydium, Orca) and return the best solution given the size, fees and low constraints.
Source: Jupiter Frontend Aggregation Example
This forces the DEX back up until it becomes too tight. With zero moat, it should be the cheapest way to win. Since they are not proprietary, backends are vulnerable to modification. We see this when pump.fun replaced Raydium as its currency with its own AMM, which affected part of Raydium’s volume.
Currently, Hyperliquid does not face this problem. Pioneering code building for perps, it’s a one-stop-shop. However, if developers switch from UI on top of HL to true routers that can send to competitive traffic, they start to resemble a smart router in traditional economics. In this example, developers can:
- Calculate the total cost: Calculate the spread/decline + buyer/producer’s fee + builder’s interest – discount + expected cost.
- Landing: Search for lots of building blocks or be turned away due to the threat of moving elsewhere.
- Capture user relationships: As sites are forced to compete to be the cheapest, the best performers offer the most revenue.
Similarly, in traditional finance, wholesalers compete with wholesalers for maximum revenue. Robinhood’s approach to Citadel Securities, Virtu, and Jane Street based on which offer the best delivery and payment options for orders.
Although competing DEXs like Drift and Ostium have included build codes, none have emerged as competitors so far. However, a major risk remains: If a site like Lighter were to include developer refunds with its zero-fee model, it would allow wallets like Phantom and Rabby to bypass Hyperliquid’s 4.5 bps fee. This would allow front-ends to capture all the fees, doubling the amount they earn per trade compared to the current Hyperliquid model.
LiquidTrading it acts as a guiding sign for the future. The Paradigm-backed terminal, which raised $7.6 million in its seed, contributed $5.6 billion to the increase in Hyperliquid. But most importantly, it also allows users to trade on Ostium and Lighter through the same interface. If larger builders follow this strategy and start operating quickly based on real estate returns instead of loyalty, Hyperliquid builder frontends could turn into a commodity for sale, directly threatening the protocol’s ability to take value.
However, there is a big difference. Spot is easy to integrate because every exchange is atomic and its properties are fungible in all places. One transaction consists of one load, and a router can divide the transaction into multiple pools. However, with perps, roles are more persistent and location specific. A BTC-PERP position on Site A may not be affected by a BTC-PERP position on Site B due to differences in policies, currency rates, lock-in engines and risk limits.
In order to thrive in different environments, the market needs one of two complex processes:
- User segmentation: Users have to save collateral in multiple places, which is not practical and makes the UX difficult.
- Main features of the brokerage: The router should act as a control unit, solving the difficult problems of credit expansion, including borders and blocking connections.
Although the lack of fungibility provides short-term protection, the terrible truth is that the front is the economic success; they will move if the competitor offers a higher margin. However, the data show that this risk exists. Although most users count on third-party connections, the majority of the volume, over 90%, still comes from the Hyperliquid front.
In addition, the HYPE token adds a storage component. Builders can have HYPE to get fee discounts, allowing them to keep their income: shipping, builder fees, and volume discounts. With this, the cost of switching to a more efficient charging system may not be necessary for the current results. Finally, the exodus of builders appears to be additive rather than cannibalistic. These are new users who enter the ecosystem through wallets and terminals, not users who change the interface.
Therefore, while the building codes provide an active vector, the expectation that Hyperliquid will remain dominant in its distribution sector is unrealistic. As the segment grows, Hyperliquid will face a tough battle to defend its lead from its competitors. However, creating an onchain orderbook for players remains the biggest technology trend, and while the front end remains healthy, incentives for developers to change are limited. However, in a fast-growing market, this is not a battle to maintain volume, but a highly competitive competition for expansion where Hyperliquid remains the dominant force.
Get news in your inbox. Check out the Blockworks blog post:
