What the fee-generating assets look like on 30 September 2026

Market-Cap-to-Fees: Spread and What It Does Not Settle

As of 30 September 2026, the market-cap-to-fees ratio (fees annualized from the trailing 30 days) ranges from 2.2x for Uniswap to 19,681x for Bitcoin, a spread of roughly four orders of magnitude. TRON sits at 110x, Solana at 231x, and Chainlink at 197x, while Zcash reaches 13,665x and NEAR Protocol 5,032x. A wide ratio indicates that market capitalization is large relative to the fee volume the network or application currently generates; it does not by itself indicate overvaluation, undervaluation, or any expected change in either variable. Fee volumes may be episodic, and a single trailing 30-day window can amplify or suppress the annualized figure materially.

Fee Retention: Protocols That Keep Versus Protocols That Pass Through

The dataset distinguishes fees paid by users from revenue retained by the protocol. TRON, Hyperliquid, Canton, Avalanche, Tether Gold, and Chainlink each report a take rate at or near 100%, meaning the figures for fees and revenue are effectively identical under their respective accounting bases. Ethereum retains 30.6% of its fees as revenue (annualized from the trailing 30 days), Sui 37.9%, and NEAR Protocol 70.0%. Uniswap, by contrast, shows a take rate of 7.6%, passing the substantial majority of its $2.52 billion annualized fees to liquidity providers rather than to a protocol treasury. Bitcoin and Zcash carry null revenue figures, meaning no protocol-level retention split is reported under their accounting bases. These structural differences mean that fee totals and revenue totals answer different questions and should not be aggregated or ranked interchangeably.

Chain-Basis Versus Application-Basis: An Accounting Boundary That Matters

Eleven assets in this dataset are measured on a chain basis (e.g., chain:Bitcoin, chain:Ethereum, chain:Solana), while Uniswap uses application:Uniswap V3, Chainlink uses application:CCIP, and Tether Gold uses application:Tether Gold. Chain-basis figures capture all fee activity settled on that network, including activity generated by applications running on top of it. Application-basis figures capture only what flows through that specific protocol. Comparing, say, Ethereum's chain-level fees with Uniswap's application-level fees does not produce a like-for-like ratio: a portion of Uniswap's fees is already embedded within Ethereum's chain total. Readers should treat these two groups separately and avoid constructing rankings that place chain and application figures in the same column without acknowledging the overlap.

Why These Are Not Equity Multiples

The single most consequential difference between a market-cap-to-fees ratio here and a price-to-revenue multiple in equity analysis is the absence of a legal claim. Holding a token does not, in general, confer a contractual right to fee flows, residual earnings, or liquidation proceeds in the way that holding a share typically does. The relationship between fee generation and token value is governed by each protocol's specific mechanism design—some burn tokens, some distribute to stakers, some accumulate in a treasury with discretionary governance—and those mechanisms are not uniform across this dataset. The multiples describe a quantitative relationship between two observable figures; they do not replicate the economic structure that gives equity multiples their interpretive meaning.