Understanding Protocol Revenue, Fees, Incentives, and Value Leakage

Hello again. In the previous lesson, you mapped the people and contracts that can change a protocol’s rules. This lesson follows the economic consequences of those rules: when someone pays a fee or sees an attractive yield, where does the value actually go?

That question is fundamental to DeFi analysis. A protocol can show high volume, high fees, and high APY while retaining little revenue, paying most rewards through token inflation, or losing meaningful value to external actors. By the end of this lesson, you should be able to turn a headline such as “this protocol generated $10 million in fees” into a clear cash-flow map.

---

First principle: these labels describe different sides of a flow

The terms in this lesson are often used as if they were interchangeable. They are not.

A user fee describes the payer: a trader, borrower, depositor, or vault user pays for a service.

Protocol revenue describes the recipient: the portion retained by the protocol, typically in a treasury, fee-collection contract, or burn mechanism.

Liquidity-provider earnings describe another recipient: the portion paid to the users who supply tradable liquidity, lending liquidity, or risk-bearing capital.

Token incentives describe the funding mechanism: tokens are distributed to encourage behavior, such as supplying capital, borrowing, voting, or trading. They need not come from user fees.

Value leakage describes value that leaves the intended users, liquidity providers, and protocol stakeholders, often through execution costs, external intermediaries, dilution, or extractable value.

So the same dollar can receive more than one label depending on the question:

• A trader’s $1 swap fee is a user fee.
• If $0.80 goes to active liquidity providers, it is also LP fee earnings.
• If $0.20 goes to the treasury, it is also protocol revenue.
• If the protocol later pays out newly minted governance tokens, that payout is a token incentive, not a share of the trader’s $1 fee.
• If poor execution lets a searcher extract additional value from the trade, that is potential value leakage from the trader’s perspective.

The practical habit is to ask three separate questions:

1. Who paid or supplied the value?
2. Who received it first?
3. What economic claim does the recipient have on it?

---

A compact vocabulary for protocol economics

Category	Core meaning	Typical source	Typical first recipient	Common mistake
User fees	Explicit payment for using a service	Traders, borrowers, vault users, issuers	LPs, protocol, affiliates, validators	Calling all fees “protocol revenue”
Protocol revenue	Fee share retained by the protocol before its expenses	A protocol take rate, reserve factor, withdrawal fee, performance fee	Treasury, fee collector, burn mechanism	Treating revenue as profit or token-holder income
LP earnings	Compensation for supplying liquidity or bearing market risk	Swap fees, borrower interest, trading fees	LPs, lenders, market makers, pool-token holders	Treating fee income as the LP’s total return
Token incentives	Token distributions meant to induce behavior	Newly issued tokens, treasury inventory, third-party grants	Suppliers, borrowers, traders, stakers, voters	Calling emissions “organic yield”
Value leakage	Value captured outside the intended economic recipients or lost through a design/friction	MEV, adverse execution, external fees, inflation, incentives sold into the market	Searchers, validators, interfaces, external service providers, new token holders	Assuming every external payment is theft or every loss is a fee

The boundary must be stated. For example, Ethereum gas paid by a swapper is not revenue for an AMM. It is an external execution cost. From the AMM protocol’s perspective, this may be a necessary cost of using the chain rather than “leakage.” From a user’s perspective, it remains part of the all-in cost of executing the swap.

---

Fees are gross activity; revenue is the protocol’s retained share

A useful metric convention comes from Token Terminal: revenue is the portion of fees a project retains through its take rate, before operating expenses and token incentives. That distinction is especially useful when comparing protocols that split user-paid fees with lenders or LPs.

Consider a hypothetical AMM swap with a $100,000 trade size and a total swap fee of $0.30\%$.

$$\$100,000\times 0.003=\$300$$

The $300 is the gross user-paid swap fee. Suppose the protocol’s fee settings direct $250 to active LPs and $50 to its fee collector.

$$\$300=\$250_{\mathrm{LP}}+\$50_{\mathrm{protocol}}$$

The correct classification is:

Flow	Classification
Trader pays $300	User fee
Active LPs receive or accrue $250	LP fee earnings
Protocol retains $50	Protocol revenue
Protocol spends $20 later on development, grants, or infrastructure	Operating expense, not a reduction in the historical gross-fee figure
Protocol later distributes governance tokens worth $80 to LPs	Token incentive, separate from the $250 fee earnings

The protocol may subsequently use the $50 of revenue in many ways: hold it in the treasury, fund operating costs, buy tokens, burn tokens, build an insurance reserve, or distribute it to some stakeholder group. Those later decisions matter, but they do not change the basic fact that the protocol’s gross revenue was $50.

A concrete AMM example: Uniswap’s fee layers

Uniswap’s documentation makes the split visible. Its basic framework distinguishes the swap fee paid on every trade from the LP and protocol portions of that fee. It also adds an important concentrated-liquidity detail: in v3 and v4, an LP earns only when its liquidity position is in range.

Two implications follow.

First, trading volume is not revenue. If a pool processes $1 billion of volume, that figure says nothing by itself about the fee rate, the LP share, the protocol take, or external routing costs.

Second, LP fee earnings are not necessarily LP profits. An LP might accrue $250 of fees but lose more than $250 from inventory rebalancing as prices move. In a concentrated-liquidity position, an LP can also be out of range and earn no fees at all. “Fees earned” is a component of return, not a complete return calculation.

---

The same separation in a lending market

The same logic applies outside exchanges.

Suppose borrowers in a lending pool pay $1,000 of interest during a period. The protocol has a $10\%$ reserve factor, meaning it retains $10\%$ of borrower interest. The remaining $90\%$ supports supplier yield.

$$\$1,000=\$900_{\mathrm{lenders}}+\$100_{\mathrm{protocol}}$$

In this case:

• The $1,000 is a user fee, paid by borrowers for access to capital.
• The $900 is lender earnings from supplying liquidity.
• The $100 is protocol revenue retained through the reserve factor.

A lending dashboard may advertise a supplier APY that includes both the $900 of interest and a separate governance-token reward. If it does, those should be shown as separate lines. Otherwise, a temporary incentive program can appear to be recurring borrower-funded income.

Aave provides an example of this distinction in Token Terminal’s project definitions: Aave’s reported revenue includes its retained portion of borrower interest, determined by reserve factors, as well as other protocol-level fee streams. The part of borrower interest paid to suppliers is not protocol revenue.

---

Token incentives: a transfer, not evidence of operating income

A token incentive is a reward for a desired action. Protocols use them to bootstrap liquidity, attract borrowers, reward trading, encourage voting, or support a new chain deployment.

The reward can be funded in several very different ways:

Incentive funding source	Who ultimately bears the cost?	What it means
Newly minted protocol tokens	Existing and future token holders through dilution, if demand does not absorb the new supply	Emissions
Tokens already held by the treasury	Treasury/token holders	Spending an existing asset reserve
Tokens purchased using protocol revenue	Protocol treasury	Revenue-funded incentive
Tokens supplied by another protocol, chain, or foundation	The external sponsor	Third-party subsidy
Revenue directly redirected to participants	The protocol or its token holders, depending on the alternative use of revenue	Revenue share, not emission

The key analytical rule is:

Do not call a return “fee yield” until you can identify the user payment that funds it.

CoinGecko’s video contrasts the early model of high emissions with an approach where participant returns draw on actual fee income. Treat “real yield” as an informal label, not a standardized accounting term: the important work is still to identify the source and recipient of each flow.

Suppose a lending market pays suppliers $900 of borrower-funded interest and adds $600 worth of newly emitted governance tokens. Its displayed return may include $1,500 of value for that period, but its sources are different:

Return component	Amount	Economic source	Durability question
Supplier interest	$900	Borrower payments	Will borrowing demand and interest rates persist?
Governance-token reward	$600	New token issuance	Will demand absorb new supply, and how long will rewards continue?
Total displayed reward value	$1,500	Combination of both	Do not describe all $1,500 as lending income

Even the stated dollar value of an incentive is uncertain until the recipient can sell it. If rewards are continuously sold, the emissions can impose persistent downward pressure on the token price. That is one way an apparent reward for liquidity providers can become value leakage borne by token holders.

---

Value leakage: costs and extraction outside the intended split

“Leakage” can become a vague catch-all unless you define the intended economic arrangement first.

For an AMM, the intended arrangement may be:

• traders receive execution;
• LPs receive the agreed share of swap fees for supplying liquidity;
• the protocol receives its specified take rate.

Anything that systematically transfers value elsewhere deserves examination. Common categories include:

Potential leakage or external cost	Who receives the value?	Why it matters
Chain gas and priority fees	Validators, sequencers, or block builders	Raises the user’s all-in cost but is not AMM revenue
Interface, aggregator, affiliate, or hook fees	Front-end operators, routers, developers	Can sit outside the core protocol fee split
MEV extraction	Searchers, builders, validators, sometimes LPs indirectly	Can worsen a user’s execution or capture value created by transaction ordering
Excessive token emissions	Recipients who sell rewards, at the expense of diluted holders	Can finance a high APY without user demand
Incentive spending with little retained activity	Short-term liquidity providers or traders	Converts treasury assets into temporary metrics
Poor routing or avoidable price impact	Arbitrageurs or better-positioned traders	Reduces execution quality even if the explicit fee is low

Not every external payment is inherently harmful. A network fee helps pay for block production and security; an aggregator may provide useful routing; a hook developer may supply a specialized service. The analytical task is not to label every non-treasury payment as bad. It is to make it visible, assess what service it purchases, and decide whether the cost is proportionate.

MEV as a form of execution leakage

A sandwich attack is a clear example of a value transfer that may not appear in a protocol’s stated fee rate. A searcher observes a pending swap, trades before it to worsen the victim’s price, allows the victim’s trade to execute at that worse price, then trades after it to close the position.

The important distinction is:

• The AMM’s published swap fee is an explicit user fee.
• The bot’s extracted profit is not necessarily an AMM fee or protocol revenue.
• The trader’s total economic cost can include the explicit swap fee, gas, ordinary price impact, market movement while waiting, and attack-related adverse execution.

The graphic illustrates the risk of a swap without sufficiently restrictive execution protection. Its numbers are illustrative, not a formula for every attack. In practice, a trade can fail if the user’s minimum-output setting is not met; that setting is a key protection against unacceptable execution, though it does not eliminate all MEV or market-risk costs.

---

Build a flow ledger, not a headline metric

When analyzing a protocol, use a short ledger for each product line: spot swapping, lending, perpetuals, vaults, stablecoin issuance, or bridging. Keep fee generation, distribution, and later spending in separate rows.

Here is a reusable template.

Step	Record	Classification test
1	The service and payer	Who paid for what: trade execution, borrowed capital, vault management, minting, or withdrawal?
2	Gross fee or interest	What was the full amount paid by the user before any splits?
3	First recipient of each share	Did it go to LPs, lenders, a protocol collector, an affiliate, a validator, or another contract?
4	Protocol take	Which share is retained by the protocol under its configured rules?
5	Supply-side compensation	Which share belongs to LPs, lenders, or market makers?
6	Incentives	Are additional rewards minted, treasury-funded, sponsor-funded, or revenue-funded?
7	External costs and extraction	What value goes to gas, routing, oracle services, MEV searchers, or other parties?
8	Net economic interpretation	After fees, incentives, market exposure, and costs, who is actually better or worse off?

A rigorous report should use language like this:

During the period, traders paid gross swap fees of $X. Active liquidity positions accrued $Y of those fees, while $Z was retained by the protocol’s fee collector. Separately, the protocol distributed $I in governance-token incentives, funded by emissions or treasury assets. Traders also bore chain fees and may have experienced execution costs outside the stated swap fee. Therefore, gross fees, protocol revenue, LP fee income, and displayed LP APY should not be treated as equivalent measures.

This wording prevents four common errors:

1. Calling gross user fees “revenue.”
2. Calling LP fee accrual “LP profit.”
3. Calling token emissions “organic yield.”
4. Ignoring costs borne outside the protocol’s treasury.

---

A fast diagnostic for advertised yield

Whenever you see an unusually high APY, break it into named components:

$$\text{Displayed return}=\text{service-fee income}+\text{token incentives}+\text{asset-price effects}-\text{costs and adverse outcomes}$$

This is not a precise return formula for every product. It is a classification discipline.

For example, a liquidity-pool position can earn swap fees, receive governance-token rewards, and still lose value because its token inventory changes as market prices move. A supplier can earn borrower interest, receive an incentive, and still face smart-contract, liquidation, or asset-price risks. The displayed APY is only one part of the position’s economics.

---

Key takeaways

A DeFi protocol’s economics become much clearer once you stop using “fees,” “revenue,” and “yield” as substitutes.

• User fees are amounts paid for a service; they can be divided among several recipients.
• Protocol revenue is the portion the protocol retains, generally before expenses and incentives.
• LP or lender earnings are supply-side compensation, often funded by user fees or borrower interest, but they are not the same as total investment return.
• Token incentives are separate transfers whose funding source must be identified: emissions, treasury assets, revenue, or an external sponsor.
• Value leakage covers economically meaningful transfers and costs outside the intended protocol, LP, and user arrangement, including MEV and dilution.
• A sound analysis starts with a gross flow, identifies the first recipient of every split, then separately records incentives, expenses, and external extraction.

In the next lesson, you will combine the protocol’s actors, privileged authorities, assets, value flows, and risks into a causal diagram.