Token Treasury Design: How to Manage Protocol Funds

Treasury design is a liquidity problem before it is a balance-sheet problem

Protocol treasuries fail for the same reason many token launches fail. The spreadsheet values assets at mark price, while the market only values what can actually be sold, borrowed against, or spent without breaking liquidity. For treasury design, the relevant question is not how much supply the protocol owns on a fully diluted basis. The relevant question is how much tradable float exists, where liquidity sits, how concentrated it is across venues, and how quickly governance can authorize action.

A treasury that is 80% native token is not diversified just because the token has a large market cap. It is a balance sheet that is structurally long its own reflexivity. If revenue slows, incentives underperform, or unlocks increase sell pressure, the treasury weakens at the same time operating needs become more urgent. That is why treasury policy belongs inside tokenomics design, not outside it.

Official DAO practice is converging toward that reality. In March 2024, a Uniswap governance RFC described the DAO treasury as worth nearly $6 billion but still “predominantly” in UNI, and argued that diversification into stable assets should be the first objective for long-term sustainability. The same RFC framed treasury UNI as economically similar to unissued equity rather than operating cash.

Size the treasury at TGE for stress, not for launch optics

Initial treasury sizing at TGE should start from burn rate, governance latency, and market depth. It should not start from an arbitrary percentage of token supply. “15% for treasury” can look generous in a launch deck and still be operationally fragile if that 15% sits in an illiquid native asset with shallow bid depth.

The Uniswap Foundation provides a clean lesson in denomination risk. On October 5, 2023, it requested $46.2 million to fund two years of operations and grants, including a 10% buffer for price volatility, and stated that it expected to return to governance when roughly one year of runway remained. The need for that buffer was not theoretical. In its earlier tranche, 2,547,002 UNI equaled $20 million at a 30-day TWAP on August 17, 2022, but by August 25, 2022, the received UNI was worth only $17.3 million after a 13.7% price decline.

The implication is straightforward. Native tokens are not treasury cash until the protocol has either converted them, hedged them, or pre-authorized realistic conversion lanes. If treasury funding is denominated in the same volatile asset whose market narrative funds the protocol, the runway is shorter than the launch model says.

At FinDaS, the practical starting point is simple: fund committed operating runway in assets the protocol can actually spend under stress. For most protocols, that means underwriting at least the next 18 to 24 months of base operating needs with stable assets or very high-liquidity reserve assets, then treating the native token reserve as strategic capital rather than payroll cash.

Asset composition should be built around operating runway, not around treasury vanity

The standard treasury failure pattern is easy to model. A protocol launches with 80%+ of its treasury in the native token, spends as if mark-to-market value is usable cash, then gets hit by a 70% token drawdown. Runway collapses. The DAO then emergency-sells into its own weakness, converting a bad market into permanent treasury damage.

A simple stress case makes the point. A $50 million treasury with 80% in native token and 20% in stables falls to $22 million after a 70% drop in the native asset. At a $1 million monthly burn, headline runway falls from 50 months to 22 months before slippage, venue fragmentation, or governance delay are even considered.

That is why the diversification argument is operational, not ideological. For real runway, many mature treasuries now anchor around 40% to 60% stable assets for operating needs, then layer in BTC, ETH, or other strategic reserve assets for medium-term optionality. The native token still belongs in treasury, but mostly as a strategic balance-sheet reserve, ecosystem incentive tool, or alignment instrument. It should not be counted one-for-one as liquid runway unless the protocol has already approved specific sale, borrow, or hedge pathways.

Aave’s treasury framework is explicit on this point. Its asset management guidelines state that the treasury should preserve capital, reduce the idiosyncratic risk of a concentrated AAVE position, and keep at least 10% of treasury value in assets that can be liquidated quickly without significant adverse price impact, citing stablecoins as the example. The same guidelines call for thorough liquidity analysis before any treasury strategy is approved.

Aave’s later governance record shows that this was not just theory. In August 2023, an Aave temperature check proposed converting $2 million of ARB into stablecoins to replenish stable reserves after a prior OTC transaction, while an earlier diversification discussion proposed converting $50 million of AAVE into stablecoins and ETH to build a stable capital pool and fund grants without pushing sell pressure onto recipients. By March 2026, an Aave funding update described the treasury as holding over $50 million in stablecoins across networks and over $100 million in non-AAVE assets.

The liquidity-structure lesson is the same across protocols. Treasury quality is defined by usable reserves, not by the optical size of native holdings. A treasury that owns 20% of token supply can still be fragile if daily sellable depth is thin, if most liquidity is concentrated on a few venues, or if unlocks are about to expand float faster than demand.

Spend policy should be denominated in stable terms and modeled in scenarios

Spend policy should be written in stable-value terms even when compensation or grants are paid partly in native token. Payroll, audits, grants, incentive campaigns, listings, legal work, and infrastructure are real-world liabilities. If the budget is framed in token units alone, the protocol quietly turns market beta into an operating assumption.

The cleanest approach is to separate spend into four buckets:

• Fixed operating burn such as core team, security, legal, and infra.
• Programmatic ecosystem spend such as grants and BD.
• Market structure spend such as liquidity incentives, market making, or protocol-owned liquidity.
• Strategic deployment capital such as treasury investments, reserve assets, or targeted acquisitions.

Each bucket should have its own funding rule. Fixed operating burn should be prefunded in stables. Programmatic spend should have quarterly caps and milestone-based release. Market structure spend should be authorized with explicit KPIs, expiry dates, and stop-loss conditions. Strategic deployment capital should face the highest approval threshold because it competes directly with runway preservation.

Uniswap’s treasury diversification policy is a practical reference. In its October 2023 funding proposal, operational funds were structured to keep six months of runway in cash, while grants funds were structured to keep a portion in stablecoins and six months of anticipated grant funding in cash. The same proposal kept a governance approval requirement for grants larger than $2 million.

Aave’s governance discussions reached a similar operational conclusion from a different direction. The 2021 diversification discussion argued that grants and investments often need stablecoins or ETH, not AAVE, because funding recipients otherwise inherit immediate sell pressure. That matters because treasury-funded grants paid in native token do not eliminate market supply. They often just delay it.

Runway modeling should be done in at least three cases: base, stress, and emergency. The stress case should haircut native token liquidity, widen slippage assumptions, and include governance delay. The emergency case should assume the protocol cannot sell size without moving its own market. A treasury model that ignores governance velocity is incomplete. If a DAO needs three weeks for forum discussion, one week for snapshot, one week for on-chain execution, and another week for operational settlement, that lag belongs in the cash forecast.

Governance should separate policy, execution, and emergency powers

Good treasury governance is not maximal direct democracy. It is a control system. Tokenholders should set policy, risk tolerance, budget ceilings, and reporting requirements. Smaller groups should execute inside those limits. Emergency powers should exist, but only with narrow scope, transparent logs, and fast post-hoc review.

Aave shows what delegated execution can look like. Its asset management guidelines call for quarterly review and monthly accounting to the community. Its 2023 financial services proposal described a non-custodial treasury management architecture with community-preapproved role-based access controls. Its GHO Liquidity Committee proposal created a 7-signer committee using a 4-of-7 SAFE, with an initial three-month budget of 406,000 GHO plus 5 ETH, and required unused funds to be returned or rolled over through governance.

Optimism uses a different institutional form but lands on similar control logic. Its capital allocation docs state that the treasury is stewarded by the Foundation, while tokenholders and other stakeholder groups oversee annual budgets through a public decision-making process. The Foundation’s operating budget is subject to Token House approval, and the documentation publishes treasury-related wallet addresses and budget flows. In 2025, Optimism also formalized a Budget Board for Seasons 8 and 9, while budget disbursement in at least some council structures used 5-of-7 multisig execution.

Transparency is not a dashboard. It is budget-to-wallet reconciliation

On-chain transparency only matters if the treasury can be reconciled from policy to wallet to realized spend. A public multisig address is useful. A public wallet plus budget approval, reporting cadence, disbursement logic, and explanation of committed versus uncommitted funds is much better.

Optimism’s capital allocation docs are strong on this front. The docs publish treasury-related addresses for Foundation-allocated and Foundation-approved OP budgets, explain that additional token movements depend on governance-approved budgets, and direct stakeholders to annual budget reports. They also disclose the broader revenue-sharing model under which Superchain member chains contribute the greater of 15% of net fee profit or 2.5% of gross transaction fees, while OP Mainnet contributes 100% of its revenue to the shared treasury.

Uniswap Foundation has also normalized periodic financial disclosure. On April 23, 2025, it published unaudited summary financials for the year ended December 31, 2024. A treasury that reports commitments, disbursements, operating expenses, and runway assumptions gives governance a basis for action before a crisis forces reactive selling.

Aave’s 2021 asset management guidelines remain one of the clearest statements of treasury reporting discipline in DeFi. They call for monthly accounting of positions and quarterly policy review. That cadence matters because treasury risk changes faster than most governance constitutions do. Liquidity moves. Venue depth moves. Revenue mix moves. Service-provider cost bases move. The treasury design has to move too.

Treasury design is a living tokenomics function

Most projects inherit a treasury structure from launch documents and never revisit it. That is usually a mistake. Treasury policy should evolve as the protocol moves from pre-revenue to fee-generating, from single-chain to multi-chain, from founder-led execution to committee-led governance, and from thin float to broader market participation.

The right treasury mix is therefore path dependent. A protocol with weak fee generation and concentrated float needs more stable runway and tighter spend caps. A protocol with recurring revenue, deep secondary liquidity, and faster governance may be able to run a larger strategic reserve in native token, ETH, or BTC. The point is not to maximize diversification for its own sake. The point is to make sure liabilities are funded by assets that remain usable when the market turns.

At FinDaS Tokenomics, treasury work sits exactly where token economy design, corporate treasury management, and on-chain governance meet. A generic chatbot can outline principles. It cannot model your specific runway needs against your specific governance velocity, float profile, unlock path, and liquidity concentration. That is why serious treasury design is not a one-time memo. It is ongoing tokenomics consulting tied to real budgets, real wallets, and real market structure.