Published on Mon Aug 24 2026 00:00:00 GMT+0000 (Coordinated Universal Time) by Jacob Cavazos

Most DeFi protocols do not explain how they make money. The fee is buried in the swap flow, the spread is hidden in the execution, the token mechanics obscure who pays what. Users swap, value moves somewhere, and the protocol’s revenue is a line item in a quarterly report that most people never read.

Orkid takes a different approach. The revenue model is a single line: 9 basis points on swap volume. No governance token extracting value. No hidden spread between quoted and executed price. No retention of surplus from price improvement. No sale of order flow data. The 9 bps fee is visible at swap time, applied to the swap amount, and that is the entire revenue surface.

This post explains the model in full — the math, why 9 bps is the number, how it compares to every major alternative, what Orkid does not do, and why this model is sustainable. The goal is transparency. If you are going to route volume through Orkid, you should know exactly where the money goes.

The 9 bps Fee: The Math

A basis point is one one-hundredth of a percent. 9 bps is 0.09%. On a swap, Orkid charges 0.09% of the swap volume as its fee.

The math is straightforward:

• On a $1,000 swap, Orkid collects $0.90.
• On a $10,000 swap, Orkid collects $9.00.
• On a $100,000 swap, Orkid collects $90.00.
• On $1,000,000 in daily swap volume, Orkid collects $900 per day.
• Over a year, $1M daily volume generates $328,500 in annual revenue.
• At $10M daily volume, annual revenue is $3,285,000.
• At $100M daily volume, annual revenue is $32,850,000.

The fee is taken on the swap amount, not on profit. There is no concept of profit in a swap — you are exchanging one asset for another. The fee is a percentage of the value that flows through the protocol. This is the same model as a payment processor (Stripe charges a percentage of transaction volume) or a stock broker (a commission per trade). It is a volume-based revenue model, not a value-extraction model.

The fee is charged to the user at the point of swap. It is included in the price you see before you confirm. There is no post-swap adjustment, no retroactive fee, no fee that appears only in settlement. The number you see is the number you pay.

Why 9 bps

The choice of 9 bps is not arbitrary. It is the result of a calculation about what a sustainable DEX aggregator fee looks like given the cost structure of running the protocol.

The costs of running Orkid are low relative to the costs of running a traditional exchange or a liquidity-pool-based DEX. Orkid does not:

• Manage liquidity pools. Orkid routes to existing DEX liquidity. It does not need to incentivize LPs with token emissions, manage pool rebalancing, or handle impermanent loss. The liquidity is already there — Uniswap, Curve, Balancer, and other DEXs have spent years building it.
• Maintain order book infrastructure. Orkid does not run an order book. It matches intents to solver bids. The infrastructure is a matching engine and solver integrations, not a high-frequency trading system.
• Fund market making. Orkid does not provide liquidity with its own capital. Solvers provide execution. Orkid provides the matching layer.
• Pay token emissions. Orkid has no governance token to emit. There is no inflationary cost to the protocol’s revenue.

Given this cost structure, the fee needs to cover: server infrastructure, solver integration maintenance, security audits (ongoing, not one-time), team compensation, and a margin that funds development. 9 bps on moderate volume covers this. At $1M daily volume, $328K annual revenue is enough to run a lean protocol team. At $10M daily volume, $3.28M annual revenue funds a serious engineering and security operation. The fee scales with the protocol’s usage, which is the correct incentive alignment — Orkid makes more money when more people use it, not when it charges each person more.

A higher fee would be more profitable per swap but would reduce volume, as users are price-sensitive and will route to cheaper alternatives. A lower fee would increase volume but may not cover costs at current scale. 9 bps is the point where the fee is low enough to not drive users to alternatives, and high enough to fund the protocol at realistic volume. It is a sustainable fee, not an extraction fee.

For context on how this compares to other aggregator fee structures, see our article on understanding DEX aggregator fees.

Comparison: How Other Protocols Make Money

To understand Orkid’s model, it helps to see how every other major DEX and aggregator generates revenue. The differences are structural, not just numerical.

Uniswap

Uniswap charges a fee of 0.05% to 1% on each swap, depending on the pool. This fee goes to liquidity providers, not to the Uniswap protocol itself. In 2024, Uniswap introduced a protocol fee that can be enabled on certain pools, diverting a portion of the LP fee to the protocol treasury. For example, on a 0.3% fee pool, the protocol might take 0.15% and the LPs take 0.15%.

From the user’s perspective, the total cost is the pool fee (0.05% to 1%) plus any MEV extraction on top. Uniswap does not provide MEV protection by default. The user pays the pool fee, pays the MEV, and the protocol may take a cut of the pool fee.

Uniswap’s model is LP-centric. The fee compensates liquidity providers for providing capital and bearing impermanent loss. The protocol’s own revenue is a secondary layer on top of the LP fee.

1inch

1inch charges a fee through a combination of surplus retention and a governance-token-based model. When 1inch executes a swap, it may retain a portion of the surplus — the difference between the quoted price and the executed price, when the executed price is better than quoted. This surplus retention is not always visible to the user. The user sees a quoted price and receives at least that price; 1inch keeps the difference between the executed price and the quote if the execution was better.

1inch also has a governance token (1INCH) used for staking and fee distribution. Users who do not hold or stake the 1INCH token do not participate in fee distribution — they pay the fee through surplus retention and receive no offsetting benefit. The user’s cost is not a single visible number. It is a combination of surplus retention, potential spread, and the opportunity cost of not participating in the token model. For users who want to know exactly what they are paying, this complexity obscures the true cost.

MetaMask Swap

MetaMask Swap charges a 0.875% markup on each swap. This is the highest fee among the options discussed here. The markup is on top of the DEX fee and gas cost. MetaMask Swap routes through third-party DEXs and liquidity sources, and the 0.875% is MetaMask’s cut for providing the routing and the swap interface.

The fee is disclosed in MetaMask’s documentation but is not always prominent in the swap interface at the moment of execution. Users who do not read the documentation may not realize they are paying nearly 1% on top of the underlying DEX fee. MetaMask Swap also does not provide MEV protection by default.

MetaMask’s model is a markup model. 0.875% is 87.5 bps — nearly 10x Orkid’s 9 bps.

CoW Swap

CoW Swap uses a batch auction model with solver competition, similar in structure to Orkid’s intent-based model. Solvers compete to execute batches of orders at the best price. CoW Swap’s revenue model includes surplus retention — when the executed price is better than the quoted price, CoW Swap may retain a portion of the surplus. CoW Swap also has a token model (COW token) that provides fee discounts and governance participation. The combination of surplus retention and token mechanics means the user’s true cost is not a single visible number. It depends on the batch outcome, the surplus generated, and whether the user holds the COW token.

CoW Swap’s model is closer to Orkid’s in architecture (solver competition, no mempool exposure) but different in revenue model (surplus retention and token mechanics vs. a flat visible fee). For users who prioritize knowing exactly what they pay, Orkid’s flat 9 bps is simpler to reason about than CoW Swap’s surplus-plus-token model.

Summary Comparison

Orkid is the only model in this list with a single flat fee, no surplus retention, no token, and MEV protection included. The 9 bps is the entire cost. There is no second layer.

What Orkid Does Not Do

Transparency requires stating not just what the revenue model is, but what it is not. Here is what Orkid does not do to generate revenue.

No Token Extraction

Orkid does not have a governance token. There is no token to emit, no token to sell to users, no token that extracts value from the protocol’s revenue. Many DeFi protocols launch a token as a revenue mechanism — the token is distributed to users or investors, and the protocol’s revenue is routed to token holders, creating a layer of value extraction that users pay for through fees, surplus retention, or inflation. Orkid has no such layer. The 9 bps fee is the only revenue. There is no token holder to pay.

This is a deliberate choice. A token creates incentives that can conflict with user interests: the protocol may be pressured to increase fees to satisfy token holders, or to prioritize token mechanics over execution quality. Without a token, Orkid’s only incentive is to provide good enough execution at a low enough fee that users keep routing volume through it. The incentive alignment is simple: more volume equals more revenue, and volume is earned by being the best option.

No Hidden Spread

Orkid does not take a spread between the quoted price and the executed price. The price you see when you sign the intent is the price you receive. There is no internal mark-up applied between quote and settlement. The 9 bps fee is applied to the swap amount and is visible in the quote.

A hidden spread is a common revenue mechanism in both CeFi and DeFi. The protocol quotes a price, executes at a better price, and keeps the difference. The user never sees the better price — they see the quote and receive the quote. This is profitable but opaque. The user cannot verify what they paid because the executed price is not surfaced.

Orkid does not do this. The solver quotes a price, the user accepts that price, and the user receives at least that price. If the solver executes better than the quoted price, the surplus goes to the user. For a full explanation of how surplus works and why Orkid does not retain it, see our article on what surplus means in DEX aggregation.

No Surplus Retention

Surplus is the difference between the price a user was quoted and the price the swap actually executed at, when the execution is better than the quote. If you are quoted 1 ETH for 3,000 USDC and the swap executes at 1 ETH for 2,990 USDC (the price moved in your favor), the 10 USDC difference is surplus.

Some aggregators retain this surplus. They quote you 3,000 USDC, execute at 2,990, and keep the 10 USDC. You receive the quoted amount and never know the execution was better. The aggregator’s revenue is the surplus, on top of any stated fee.

Orkid does not retain surplus. If the execution is better than the quoted price, the user receives the better price. The 9 bps fee is calculated on the swap amount, not on the surplus. This means Orkid’s revenue is predictable and visible — it is always 9 bps of the swap volume — and the user captures any price improvement that the solver achieves.

This is a meaningful difference. Surplus retention can add 10-50 bps of hidden cost on top of a stated fee, depending on volatility and execution quality. Orkid’s model ensures the user pays 9 bps and captures everything else.

No Data Selling

Orkid does not sell order flow data, swap data, or user behavior data to third parties. Some protocols generate revenue by selling data — order flow to market makers, swap patterns to analytics firms, or user behavior to advertising partners. This monetizes the user’s activity without explicit knowledge.

Orkid does not have a data monetization layer. The 9 bps fee is the revenue. If Orkid’s revenue were partially from data sales, the incentive would be to maximize data collection, which conflicts with user privacy. By relying solely on the swap fee, Orkid’s incentive is to maximize swap volume, which aligns with providing good execution — not with harvesting data.

Why 9 bps Is Sustainable

The question of sustainability is whether the 9 bps fee can fund the protocol indefinitely without increasing. The answer depends on the cost structure and the volume trajectory.

Orkid’s costs are fixed-to-linear, not quadratic. Server infrastructure scales with volume but is a small fraction of revenue. Security audits are a periodic fixed cost. Team compensation is a fixed cost that grows slowly with headcount. The marginal cost of an additional swap is near zero — the infrastructure is already running, the solver integrations are already maintained. Each additional dollar of volume costs Orkid almost nothing to process, and 9 bps of that dollar is revenue.

This is the opposite of a liquidity-pool model, where scaling requires more liquidity, which requires more incentives, which requires more token emissions or higher fees. Orkid does not need to fund liquidity. The liquidity exists on the DEXs Orkid routes to.

At current volume levels, 9 bps funds the protocol. As volume grows, the fixed costs (team, audits, infrastructure) are amortized over more swaps, and the margin per swap increases. This means Orkid becomes more profitable as it scales, without needing to raise the fee. The fee can stay at 9 bps indefinitely.

The risk to sustainability is not that 9 bps is too low — it is that a competitor enters with a lower fee. If a competitor offers 5 bps with comparable execution quality, Orkid faces pressure to lower its fee. This is the market functioning correctly. The defense is not a moat — it is execution quality and trust. If Orkid provides better execution (less slippage, better routing, faster settlement) than a 5 bps competitor, the total cost to the user may be lower at 9 bps despite the higher fee, because execution quality dominates the fee at moderate swap sizes. If it does not, Orkid lowers the fee. The fee is a lever, not a commitment.

The Long-Term Vision: Volume Scales, Not Fee Extraction

The long-term vision for Orkid’s revenue is simple: volume-based revenue scales with adoption, not with fee extraction. The protocol makes more money when more people use it, not when it charges each person more. This is the payment processor model applied to DeFi — Stripe does not raise its percentage as it grows; it processes more transactions. Visa does not increase its interchange fee to grow revenue; it increases the number of cards and merchants.

Orkid’s model is the same. The 9 bps fee is fixed. Revenue growth comes from volume growth. Volume growth comes from being the best execution venue — the lowest total cost (fee plus slippage plus MEV protection) for the user. The incentive is to improve execution, not to increase extraction.

This is contrasted with the token-extraction model, where revenue growth comes from increasing the value captured per swap — higher fees, more surplus retention, more token mechanisms. That model grows by taking more from each user. Orkid’s model grows by serving more users. In the extraction model, existing users pay for growth. In the volume model, new users bring growth.

For Orkid, this means the long-term revenue curve is a function of adoption. If Orkid processes 1% of DEX swap volume on Base, the revenue is one number. If it processes 10%, the revenue is 10x that number, at the same 9 bps. The fee does not change. The protocol’s job is to earn that volume by being the best option.

For a related discussion of Orkid’s safety model and why the non-custodial architecture means Orkid cannot lose your funds even if the protocol fails, see our article on is Orkid safe. For background on the MEV protection that Orkid provides, see our primer on MEV and how to protect against it. For an explanation of the intent-based execution model that underpins Orkid’s solver competition, see our article on intent-based swap execution.

Frequently Asked Questions

How does Orkid make money?

Orkid charges a flat fee of 9 basis points (0.09%) on swap volume. On a $10,000 swap, Orkid collects $9. On $1 million in daily volume, Orkid collects $900 per day, or approximately $328,500 per year. The fee is the only revenue source. Orkid does not have a governance token, does not retain surplus, does not take a hidden spread, and does not sell order flow or user data. The 9 bps fee is visible at swap time and is the entire cost of using the protocol.

Why does Orkid charge 9 bps?

9 bps is the point where the fee is low enough that it does not drive price-sensitive users to alternatives, and high enough to fund the protocol at realistic volume levels. Orkid’s cost structure is low — it does not manage liquidity pools, does not fund market making, and does not pay token emissions. At $1M daily volume, 9 bps generates $328K annually, which funds a lean protocol team. At $10M daily volume, it generates $3.28M. The fee scales with usage rather than with extraction.

Does Orkid keep the surplus from swaps?

No. Surplus — the difference between the quoted price and the executed price when execution is better than the quote — goes to the user. If you are quoted 3,000 USDC for 1 ETH and the swap executes at 2,990 USDC, you receive the 10 USDC difference. Orkid’s 9 bps fee is calculated on the swap amount, not on the surplus. Surplus retention adds hidden cost that the user cannot verify, while a flat fee is visible and predictable. Orkid’s revenue is always 9 bps of volume, regardless of execution quality.

Is Orkid’s 9 bps fee sustainable?

Yes. Orkid’s costs are fixed-to-linear and do not scale with the liquidity it accesses. The marginal cost of processing an additional swap is near zero. As volume grows, fixed costs are amortized over more swaps, increasing margin per swap without raising the fee. The fee can remain at 9 bps indefinitely. The risk to sustainability is competitive pressure from lower-fee alternatives, not cost structure. The defense is execution quality — if Orkid provides better routing and less slippage than a cheaper competitor, the total cost to the user may be lower despite the higher fee.

How does Orkid compare to Uniswap’s fee model?

Uniswap charges 0.05% to 1% per swap, paid to liquidity providers, plus an optional protocol fee that diverts a portion to the Uniswap treasury. The user pays the pool fee plus any MEV extraction (Uniswap does not provide MEV protection by default). Orkid charges a flat 9 bps on volume, includes MEV protection through intent-based execution, and does not route revenue to LPs or a token treasury. Uniswap’s model is LP-centric — the fee compensates liquidity providers for capital and impermanent loss. Orkid’s model is volume-centric — the fee compensates the protocol for matching and execution. On a like-for-like swap, Orkid’s total cost (9 bps fee plus zero MEV) is lower than Uniswap’s total cost (pool fee plus MEV extraction) on most pairs.

Written by Jacob Cavazos