Application revenue drives crypto token buybacks and burns

Application revenue and token-supply reduction are moving closer to the center of crypto valuation debates, as income has shifted away from execution-layer infrastructure and toward user-facing products, according to Austin Barack, founder and managing partner of Relayer Capital. Barack said applications now account for roughly two-thirds of industry revenue, compared with more than 95% previously generated by execution layers, and he expects applications could eventually capture more than 90%.

That change is pushing some liquid-token managers to examine projects through a framework closer to equity analysis: revenue growth, the share of revenue directed to token buybacks or burns, and the market value assigned to those distributions. The approach remains imperfect. Tokenholders generally do not have the contractual residual claims held by shareholders, and a protocol can alter a buyback or burn policy at any point. Yet revenue-funded supply reductions give traders a more measurable basis for comparing tokens than narratives alone.

Barack said Relayer Capital, founded about two and a half years ago, now directs about 95% of its work toward liquid tokens. Its focus is concentrated in Crypto×AI, round-the-clock trading infrastructure, and asset tokenization.

Revenue growth is being weighed against token valuations

Relayer’s screening process compares a project’s revenue growth with its existing token valuation. Barack argued that an extended period of valuation compression has made it easier to distinguish products with durable usage and growing income from tokens whose market values remain detached from activity.

The central calculation resembles a buyback yield or a price-to-earnings-style comparison. A protocol that spends $30 million annually buying back tokens, for example, can be judged against its circulating market capitalization. A lower implied buyback multiple may suggest the market is assigning less value to each dollar directed toward token repurchases.

The comparison has limits beyond the lack of formal tokenholder rights. A burn is not always equivalent to a cash distribution, and its effect depends on issuance schedules, unlocks, treasury sales, liquidity, and whether the underlying revenue can persist through weaker market conditions. Revenue also varies sharply across business models. Trading fees, subscriptions, lending income, and meme-token issuance activity carry very different durability profiles.

Venice projections depend on expanding burn coverage

Venice’s VVV token was among the examples Barack used to illustrate the upside and uncertainty in revenue-linked burns. He said the privacy-focused AI project generates most of its revenue from paid subscriptions and purchases of additional compute credits.

Venice has two stated burn paths for VVV. First-time subscription purchases can trigger burns that vary by plan tier. A separate mechanism allocates about 5% of spending on compute credits toward buying and burning VVV.

Barack estimated Venice’s annualized revenue run rate at approximately $107 million in August 2026, alongside annualized VVV burns of about $8.3 million. His 2027 model projected $336 million in revenue and $70 million in burns. Applying a 50-times buyback multiple to that burn figure produced an illustrative token valuation of roughly $3.5 billion, or about $43.90 per VVV, compared with about $16 at the time of the discussion.

Those figures rest on several assumptions. More than $29 million of the projected $70 million burn total was linked to Minds, a product that had not formally launched, meaning over 40% of the modelled burn depends on a new business line. The forecast also assumed burn eligibility would expand to subscription renewals and that the credit-revenue burn allocation would rise from 5% to 10% by 2027. Neither change was presented as a fixed commitment.

Venice had previously raised $65 million, according to Barack. External funding can give an early-stage application room to build products and acquire users, while revenue-funded burns can reduce supply without requiring the treasury to sell additional tokens. The distinction becomes more consequential when projections depend on future product launches rather than established revenue.

Pump.fun and Hyperliquid show different market pricing

Barack placed Pump.fun and Hyperliquid at opposite ends of the buyback-multiple range. He said Pump.fun was trading at an implied buyback multiple of about five times, while Hyperliquid and Lighter were valued nearer 30 to 40 times.

Pump.fun’s revenue is tied to meme-coin creation and trading, a category frequently questioned for its ability to retain demand through changing market conditions. Barack said the platform’s revenue had nevertheless remained resilient for more than two years. Its buyback policy has changed: after previously directing all revenue toward buybacks, the platform planned to use 50% of revenue for buybacks over the following 12 months. The longer-term allocation remained undecided.

Hyperliquid’s higher valuation, in Barack’s view, reflects its position in perpetual futures trading and its effort to broaden the types of markets available through HIP-3. The expansion includes products tied to stocks, commodities, indexes, and private-company contracts.

He said real-world-asset-related markets were producing substantial volumes but had not yet translated into comparable revenue growth, leaving crypto trading as the core profit source. That concentration creates a powerful but cyclical mechanism: rising market activity can lift fees and increase token buybacks, while a declining trading environment would reduce both.

Barack also cited Hyperliquid’s reported $6.86 billion in total value locked and $172 billion in monthly volume. Such figures indicate the scale of capital and trading activity on the platform, though volume alone does not determine how much revenue ultimately reaches buybacks.

ether.fi’s banking products alter its revenue profile

Ether.fi offers a separate example of how a project’s business mix can outgrow the category traders initially associate with it. Originally known for liquid restaking, ether.fi now earns more than 65% of its business revenue from Neo Bank products, according to figures cited by Barack. Those products include credit-card transaction revenue and borrowing income secured by users’ account assets. Yield and staking-related operations account for the remaining roughly 35%.

Barack said daily credit-card spending had increased from about $300,000 a year earlier to between $3 million and $4 million. The platform had also moved its card operation to Optimism and was managing more than 70,000 active cards, according to the discussion.

Borrowing interest accounted for only about 4% of ether.fi revenue, Barack said, compared with an estimated 60% to 70% revenue contribution from lending at Nubank. The comparison points to the distance between ether.fi’s current business and a mature lending-heavy financial platform, while also showing why a shift toward borrowing could materially alter its revenue mix.

A valuation scenario cited by Barack assumed about $30 million in ETHFI buybacks over the next 12 months and applied a 30-times multiple, producing an implied token price above $1, roughly double the level discussed at the time. A separate model estimated $21 million in buybacks; the difference came largely from assumptions about growth.

Revenue-funded token reductions can create a valuation reference point, but they do not isolate tokens from crypto’s wider liquidity cycle. Trading activity, demand for leverage, major-asset price movements, and changes in protocol policy can quickly alter both the cash available for buybacks and the multiple traders are willing to assign to it.