Hyperliquid Retail Trading: $280M in Wallet Losses
See how six retail apps brought traders to Hyperliquid—and why fees, trading losses and weak retention raise questions about sustainable growth.

Hyperliquid Retail Trading
The $280 Million Question Behind Hyperliquid's Retail Growth
Retail apps have become a fast-growing entry point into Hyperliquid's perpetual futures. Phantom, MetaMask, Rabby, Trust Wallet, Invo, and fomo connect users to the exchange through builder codes - so the apps can earn a fee when users trade. That setup can bring in new wallets, more trading volume, and more fee revenue for Hyperliquid. But it doesn't automatically mean the users stick around or end up better off.
An analysis tracking nearly 400,000 wallets linked to these six apps finds that those wallets are down $280 million all-time across Hyperliquid. The number needs context. About $126 million was lost trading through the apps from November 10, 2025 onward. The rest of the losses happened before that date or through Hyperliquid activity outside the app connections. In the more recent app-linked losses, fees account for a bit more than half.
Taken together, the numbers point to a growth channel with two very different outcomes: acquisition is speeding up, but many traders lose money and don't stay. Here's what the data says about who these traders are, what they trade, how the costs add up, and what retention means for whether app-driven DeFi adoption can be durable.
What the $280 million figure tells us
The headline number is eye-catching, but it's also easy to misread. The analysis looks at nearly 400,000 wallets that traded Hyperliquid perpetuals through those six retail apps. Across all of Hyperliquid, those wallets are down $280 million all-time. That's not the same thing as saying the apps alone caused $280 million in losses.
A tighter comparison is the $126 million lost through the apps since November 10, 2025. The rest of the all-time losses came before November or from Hyperliquid activity that wasn't routed through those app connections. Keeping those periods separate matters, because attribution drives both how we judge the user experience and how we think about the economics of the acquisition channel.
Wallet losses are not a complete account of user outcomes
A wallet is an address, not necessarily a single real-world person. One user can control multiple wallets, and a wallet's trading history won't capture the person's full financial situation or activity on other platforms. So the analysis is useful for spotting patterns in the tracked behavior - but it shouldn't be treated like a full survey of individual traders.
There's also a key distinction between trading results and fee costs. The reported losses include both outcomes from trades and fees. In the period described, fees make up slightly more than half of the $126 million lost through the apps. That pushes the question beyond "did traders pick the right direction?" It also raises "how often did they trade, what did they pay to trade, and did those costs fit the size and risk of their account?"
Why the distinction matters for growth
Builder codes give an app a reason to integrate with Hyperliquid: the app earns a cut of trading activity, while Hyperliquid gets access to a distribution channel. That's a real growth mechanism. The harder part is what it implies for users - whether they're succeeding, and whether they'll come back.
For an exchange, volume and fees are measurable quickly. User outcomes and retention take longer. If a channel brings in lots of accounts that trade briefly, rack up costs, and leave, the acquisition numbers can look strong even if the relationship isn't healthy. The data doesn't prove every app user has the same experience, but it does underline a basic question that any app-led trading model should answer: are users being brought into a product they can use sustainably?
Who is trading through the apps
The tracked group is mostly small retail accounts. At five of the six apps, the typical wallet deposited between $30 and $60. Rabby stands out: its user base is smaller, its accounts are larger on average, and its users are the most likely among these app groups to be profitable.
That matters because the same trading costs hit differently at different account sizes. A fee that feels small on a $100 order can be a meaningful chunk of a small account - especially if the trader repeats the same pattern many times. Low balances also leave less room for adverse moves and liquidation risk before the account is depleted.
Familiar assets attract the most activity
BTC accounts for close to half of the volume routed through the apps. That concentration isn't surprising - Bitcoin is widely known, has deep markets, and is often the first asset many newer crypto traders touch. But familiarity doesn't automatically mean suitability for frequent leveraged trading. Perpetual futures can amplify both gains and losses, and frequent entries and exits can turn fees into a recurring burden.
The second-largest category has shifted toward HIP-3 markets, including assets like oil, silver, and gold. The analysis says these markets barely traded through the apps before January, then grew to become the second-largest category. That's a change in what retail traders are getting access to through crypto apps: not just crypto assets, but commodity-linked markets too.
This expands the product - but it can also complicate the user experience. A trader who's comfortable with BTC may not understand the specific dynamics, trading hours, or risks of a commodity-linked contract. Easy access can reduce friction to placing a trade; it can't replace learning what you're actually trading.
Account size changes the meaning of activity
A small deposit paired with fast turnover can produce impressive volume without reflecting a large, durable customer relationship. Invo users, for example, deposited $12.4 million and traded $5.8 billion. The gap between deposits and traded volume reflects repeated use of capital through trading - not $5.8 billion of brand-new money entering users' accounts.
That distinction helps when you're looking at product metrics. Deposits, volume, and active wallets answer different questions:
- Deposits show how much capital users put into the app-linked activity.
- Trading volume measures how much notional value moved through the markets.
- Active wallets show how many addresses traded during a period.
- Retention shows whether users came back over time.
- Net outcomes after fees give a clearer view of the financial experience within the measured activity.
No single metric tells the full story. A platform can grow volume quickly while average account balances stay small and retention remains weak. Separating these measures is a practical way to tell whether "growth" is driven by lasting adoption or rapid turnover.
How losses and fees build up
The analysis points to two sources of losses: trading outcomes and fees. For the $126 million lost through the apps since November 10, 2025, fees account for slightly more than half. That doesn't mean trading skill or market conditions are irrelevant. It means transaction costs were a major driver of the observed result - not a side detail.
Each app charges its own builder fee on top of Hyperliquid's base trading fee. For a $100 order, the combined cost ranges from 6.5 cents to 14.5 cents depending on the app. On any one trade, that can look minor. The issue is repetition: if a trader keeps turning over a small balance, the fee hits again and again - whether the trade ends up profitable or not.
Turnover makes a small fee consequential
Imagine a trader with a modest account who makes frequent trades. The fee is calculated on the value traded, not just the amount originally deposited. If the same capital gets reused, total notional volume can end up many times larger than the deposit. In that case, costs stack up through turnover.
That's why Invo's reported $12.4 million in deposits alongside $5.8 billion in trading volume is useful. It shows how quickly repeated trading can generate a large volume figure relative to the capital users actually put in. It doesn't, by itself, tell you what any individual trader paid or lost - but it helps explain why per-trade charges can add up across a high-turnover user base.
So the practical question for an app isn't just whether its fee is competitive on one order. It's whether users can understand the full cost of frequent trading, and how the app's fee combines with Hyperliquid's base fee. Clear pricing matters even more when the target users have small balances and are likely to trade repeatedly from a phone.
Market choice affects the cost profile
BTC is the most traded market in the analysis, and also the market where users lose the most. On BTC, fees alone exceeded what users made on the trades they closed. That's a specific finding about the tracked activity, and it shouldn't be generalized to every BTC trader or every market period.
HIP-3 and smaller markets also cost users more per dollar traded than BTC and ETH combined, according to the analysis. That creates layered risk: unfamiliar or thinner markets may behave differently, and the fee burden per dollar can be higher. Adding more markets can broaden what the product offers, but broader access doesn't automatically translate into better outcomes - especially for users who don't fully understand the costs and risks.
For anyone deciding whether to trade, the relevant comparison is broader than the advertised app fee. Users need to consider combined app + exchange charges, expected turnover, leverage, market conditions, and the real possibility of liquidation. A platform can make an instrument easy to access without making it easy to use well.
Growth is clear; retention is not
Across the six apps, monthly active traders more than tripled from November to September. Invo and fomo drove much of the increase, going from almost no traders to tens of thousands within a few months. As acquisition channels, these apps have shown they can bring a meaningful number of users into Hyperliquid markets.
Retention tells a different story. About half of the active traders in each month are trading through the apps for the first time. Only around one in seven newly active traders is still trading three months later. Put simply: the growth described in the analysis relies heavily on continually onboarding new wallets, not on a large share of users returning over time.
Acquisition and retention measure different things
A new trader is a useful acquisition outcome. A returning trader is evidence that the product actually meets a need. Neither metric alone proves a business is healthy, but the gap between them helps explain the model.
If you have high acquisition and low retention, the platform has to keep finding new users just to sustain activity. That can work when distribution is cheap and the audience keeps expanding. It gets harder when acquisition slows, competing apps offer similar access, or users decide the costs and risks outweigh the benefits.
Low retention doesn't automatically mean a product failure. Some people trade for a short time by choice. High-risk products naturally have different usage patterns than savings accounts or daily-use apps. The more telling point here is the combination: weak three-month retention alongside reported losses and substantial fee drag. Those signals deserve more weight than any single headline "growth" number.
What the data can and cannot establish
These are observational results. They describe wallet activity and outcomes within the tracked cohort, but they don't prove why someone stopped trading, whether fees drove that decision, or whether every app delivers the same experience. The analysis also can't show how these outcomes compare with every other route into Hyperliquid.
Still, the pattern is hard to ignore. When roughly half of monthly actives are new and only about one in seven new users is active three months later, ongoing onboarding is doing most of the work. If many of the tracked users are small accounts and fees account for more than half of recent app-linked losses, then the economics of repeat trading deserve closer scrutiny.
This is the trade-off in the distribution model: apps can grow quickly by making access easy, but the long-term quality depends on what happens after the first trade. A healthy channel should be judged not just by how many users it brings in, but by whether people understand what they're using, return voluntarily, and can make informed decisions about costs.
What sustainable app-led adoption would require
On the surface, builder codes are doing what they were meant to do: connect consumer apps to Hyperliquid, bring in new traders, and generate volume and fees. Those outcomes are real. The concern is that users entering through this route are, on aggregate, losing substantial amounts in the tracked activity - fees make up a large share of recent losses - and many users don't return after a few months.
If this model is going to last, it can't rely on acquisition alone. Apps and exchanges have different roles, but the core product questions overlap: Are trading costs clearly disclosed? Can users understand when app fees are added on top of exchange fees? Is leverage risk made clear before a trade? And do users have a way to see how turnover and fees affect their account over time?
Product design should make costs legible
Cost transparency isn't a substitute for risk controls, and it won't magically improve trading outcomes. But users can't make informed decisions about frequent trading if the total cost is difficult to find or hard to understand. Showing the combined cost in context - rather than treating each charge as a small, isolated line item - would help users judge what those fees mean relative to their account size and trading habits.
Similarly, if a product offers access to more markets, it should help users tell those instruments apart. BTC, ETH, and commodity-linked markets aren't interchangeable just because they appear in the same interface. Good product design makes the important differences visible instead of assuming users will learn them only after they've already placed trades.
Retention should be interpreted alongside user outcomes
Retention is a business metric, but it shouldn't be treated as an objective "at any cost." Encouraging people to trade more often just to prop up activity numbers is the wrong response to weak retention. The better goal is useful, informed repeat usage: users come back because the product serves them, because they understand what they're trading, and because they can see the costs involved.
For teams building consumer access to DeFi, a more balanced scorecard might include:
- New users and the cost of acquiring them.
- Repeat activity at one, three, and six months.
- Fees paid relative to deposits and trading volume.
- User understanding of leverage and liquidation risk.
- The share of volume coming from existing users versus new wallets.
- Complaints, support requests, and reasons users stop trading.
None of these measures removes market risk. But they can show whether growth is creating a lasting product relationship - or depending on a constant stream of first-time traders.
The source analysis expects more consumer apps to integrate with Hyperliquid through builder codes. That seems plausible: integrations can make decentralized trading easier to reach. But the apps that stick will need more than a one-time wave of wallets. They'll need to keep users without relying on unclear costs or high-frequency activity as the main engine of growth.
Conclusion
The app integrations have worked for distribution: they bring retail wallets, trading volume, and fee revenue to Hyperliquid. The tougher question is whether that growth can last - especially if many tracked accounts are small, recent app-linked losses include significant fee costs, and only a minority of new traders are still active after three months.
This won't be answered by adding another acquisition channel. It depends on whether costs and risks are understandable, whether retention is measured alongside user outcomes, and whether the product remains useful after the first trade. If you're assessing app-led DeFi growth, start by separating new wallets and volume from repeat usage and net outcomes. That's where the gap between short-term activity and durable adoption becomes clear.
For the underlying figures and methodology, read the original analysis.