Take rate in crypto protocols

Gross fees look impressive. Take rate reveals business quality. If a protocol collects $100M in fees but retains 1%, that's a $1M business. Investors who ignore take rate systematically overpay.

Key takeaways

Take rate measures what percentage of fees a protocol actually keeps versus what it passes through
Gross fees are misleading; they include payments to LPs, stakers, and other required parties
Take rates vary dramatically by protocol type: DEXs typically 5-20%, lending 10-30%, perps 20-50%
Low take rates aren't inherently bad if volume is high, but they limit pricing power and sustainability
Incentives can artificially inflate volume while depressing take rates, creating unsustainable economics

What "Take Rate" Means Onchain

Take rate is simple: what the protocol keeps divided by what passes through. A DEX collects 0.3% swap fees. LPs receive 0.25%. The protocol retains 0.05%. The take rate is 16.7% (0.05 / 0.3).

This matters because it determines monetization potential. Two protocols can have identical fee volume with 10x different take rates. One earns $10M annually. The other earns $1M. The fee charts look similar. The business quality is completely different.

Take rate reveals pricing power and value capture. High take rates mean the protocol extracts substantial value from each transaction. Low take rates mean the protocol is a pass-through entity with minimal retention. Volume obscures this. Take rate clarifies it.

Why Gross Fees Are Misleading

Dashboards often report "total fees" or "total revenue" when they mean gross fees collected. This number includes everything: LP shares, validator tips, staker rewards, protocol retention. It's the numerator. It's not what the protocol keeps.

Gross fees create false comparisons. Protocol A generates $500M in fees with 5% take rate. Protocol B generates $100M in fees with 30% take rate. Protocol A retains $25M. Protocol B retains $30M. Gross fees say A is larger. Take rate reveals B captures more value.

This confusion is systematic. Marketing departments highlight gross fees because they're larger numbers. Investors unfamiliar with crypto economics assume fees equal revenue. They don't. Take rate is the conversion factor that separates appearance from reality.

Take Rates Across Models

DEX

DEXs typically have low take rates. Uniswap v2 had 0% (all fees to LPs). Uniswap v3 introduced optional protocol fees of 10-25% of the LP fee, activated by governance. Most pools still send 100% to LPs.

Curve takes 50% of trading fees for veCRV holders and the DAO. If a pool charges 0.04% fees, Curve retains approximately 0.02%. That's a 50% take rate on fees, but fees are small relative to volume.

Low DEX take rates reflect competitive dynamics. LPs can move liquidity to higher-yielding venues. Raising protocol take rate means reducing LP yields, which reduces liquidity, which widens spreads, which loses users. It's a tight equilibrium.

Lending

Lending protocols have moderate take rates. Aave's reserve factor (protocol take rate) varies by asset, typically 10-35%. If a market has 5% borrow rate and 3% supply rate, the 2% spread has an implicit take rate component that goes to the protocol and insurance.

Compound has similar dynamics. The protocol retains a percentage of the interest spread. This percentage can be adjusted by governance but faces the same competitive constraints as DEXs. Too high and capital moves elsewhere.

Lending take rates balance protocol sustainability with competitive rates. Higher take rates fund reserves and insurance. Lower take rates attract more deposits and borrowers. The optimal rate depends on market position and risk profile.

Perps

Perpetual futures protocols can sustain higher take rates. GMX charges 0.1% open/close fees plus funding rates. A significant portion flows to GLP providers, but the protocol retains 30% of fees for GMX stakers. Effective take rate on fees is around 30%.

dYdX has trading fees of 0.05-0.20% depending on tier. Market makers receive rebates. The protocol retains the net after rebates. Take rates vary but are generally higher than DEX spot trading because perpetuals have more complex risk management.

Higher perps take rates reflect value-add. Perpetuals provide leverage, funding rates, and capital efficiency. Users pay for these features. The protocol can charge more without losing users to alternatives.

Liquid Staking (LST)

LST protocols have transparent take rates. Lido charges 10% of staking rewards. If staking yields 4%, validators get 3.6%, Lido gets 0.4%. The take rate is 10% of gross staking yield.

Rocket Pool has a different structure but similar magnitude. The protocol takes a percentage of rewards to fund node operators and the protocol. Take rates in LST are explicit and comparable.

LST take rates are sticky because switching costs are high. Unstaking requires time. Liquidity for derivative tokens varies. Once users deposit, the protocol can maintain take rates without immediate competitive pressure. This enables sustainable revenue capture.

L2 Sequencer

L2s have variable take rates depending on L1 costs. An L2 collects transaction fees from users. It pays data availability costs to Ethereum L1. The difference is gross sequencer revenue. The take rate is this difference divided by user fees collected.

When L1 gas is cheap, L2 take rates can be 80%+. When L1 gas spikes, take rates compress or go negative if the L2 subsidizes users. This makes L2 revenue volatile and dependent on exogenous factors.

Future L2 take rates depend on blobs and data availability improvements. Lower L1 costs should expand L2 take rates. But competition between L2s may force fee reductions that compress take rates despite lower costs.

Take Rate vs Sustainability

Low take rates require high volume to generate material revenue. A 2% take rate on $1B annual volume is $20M revenue. On $100M volume, it's $2M. The business only works at scale. Pre-scale, it's unprofitable.

High take rates enable profitability at lower volume. A 30% take rate on $100M volume is $30M revenue. This creates faster paths to sustainability. But it also attracts competition and risks alienating users if take rates rise too high.

The optimal take rate balances current profitability with future market share. Too low and the protocol never profits even at scale. Too high and competitors with lower take rates capture market share. Protocol teams must find the equilibrium.

Governance complicates this. Tokenholders want higher take rates to maximize revenue. Users want lower take rates. LPs want higher yields. Balancing these incentives through governance is challenging. Protocols often oscillate between extremes.

Incentives and Take-Rate Distortion

Emissions can inflate volume while depressing effective take rates. A DEX offers 50% APY in token rewards to LPs. Volume increases because yields are high. But gross fees don't account for the cost of those emissions. The net take rate (revenue minus emissions) is deeply negative.

Calculate true take rate by including emissions as a cost. If a protocol retains $5M from fees but distributes $20M in tokens, the net take rate is -$15M on whatever the fee base was. The protocol is paying for volume, not earning from it.

Sustainable take rates exclude subsidies. What would the take rate be if emissions stopped? If volume collapses when rewards end, the take rate during the rewards period was artificial. Real take rate emerges only after incentives taper.

What Good vs Bad Take Rates Look Like

Good take rates are sustainable, defensible, and scale with usage. They cover costs, fund development, and leave room for tokenholder distributions. They don't rely on temporary competitive advantages or user lock-in that might erode.

Bad take rates are either unsustainably low or artificially high. Unsustainably low means the protocol earns nothing even at scale. Artificially high means they're defended by lock-in that will eventually break or regulation that will change.

Industry benchmarks help. DEXs at 5-15% take rates are reasonable. Lending at 15-30% is sustainable. Perps at 25-50% is defensible. LSTs at 5-10% is standard. Deviations require explanation. Extremely high or low take rates signal either innovation or unsustainability.

Protocol Type	Typical Take Rate	Revenue Drivers
DEX	5-20%	Volume, competitive LP yields
Lending	15-30%	Interest spread, utilization
Perps	25-50%	Trading fees, funding rates
Liquid Staking	5-10%	Staking rewards, validator performance
L2 Sequencer	Variable (0-90%)	User fees minus L1 costs

See live data

Links open DefiLlama or other external sources.

Related Concepts

Take rate analysis requires understanding protocol revenue mechanics:

Protocol revenue: Learn how retained revenue relates to take rate
Fees vs revenue vs profit: Why gross fees differ from actual protocol retention
DeFi business models: How different protocol types generate revenue
Cost of funds in DeFi: Understanding what drives the gap between fees and revenue
Onchain profit: How take rate affects ultimate profitability
TVL vs revenue: Why capital efficiency matters more than locked value

FAQ

Is a higher take rate always better?

Not necessarily. Higher take rates generate more revenue per transaction but can reduce volume if users find cheaper alternatives. The optimal take rate balances revenue per transaction with total transaction volume. Too high loses market share. Too low never achieves profitability.

Why don't all protocols maximize their take rate?

Competition limits take rates. If Protocol A charges 0.3% and Protocol B charges 0.1% for the same service, users and liquidity providers migrate to B. Protocols must balance revenue extraction with remaining competitive. Market power determines how much pricing flexibility exists.

How do I calculate take rate from public data?

Divide protocol retained revenue by gross fees collected. DefiLlama reports both fees and revenue for many protocols. Take rate is revenue/fees expressed as a percentage. If fees are $100M and revenue is $20M, take rate is 20%.

Can take rate change over time?

Yes, through governance or market conditions. Protocols can vote to increase or decrease take rates. Competitive pressure can force reductions. Cost changes (like L1 gas for L2s) can alter effective take rates. Monitor take rate trends, not just snapshots.

What's the relationship between take rate and emissions?

Emissions can create negative net take rates. If a protocol retains 10% of fees as revenue but spends 30% of fees on token incentives, the net take rate is -20%. True profitability requires revenues (take rate × volume) to exceed emission costs.

Do users care about protocol take rates?

Indirectly. Users care about total cost (fees paid). Whether the protocol keeps 10% or 30% of that fee matters less than the absolute fee amount. But take rate determines if the protocol can sustain service quality long-term. Unsustainably low take rates lead to protocol failure.

Which protocol type has the most pricing power?

It varies, but generally perps and LSTs have more pricing power than DEXs or lending. Perps offer unique leverage features. LSTs have switching costs. DEXs face intense competition on fees. Lending is moderately competitive. Pricing power correlates with differentiation and switching costs.

Cite this definition

Take rate is the percentage of gross fees or transaction value that a protocol retains as revenue after required pass-throughs to LPs, stakers, and other capital providers, revealing true business economics better than gross fee metrics that include all payments flowing through the system.