How to Structure DAO Governance Tokens for Real Utility

Governance tokens have utility only when they control enforceable rights

Governance tokens have real utility only when they let holders or their delegates make binding decisions over scarce protocol rights: upgrades, treasury outflows, emissions, risk parameters, or the appointment and removal of decision-makers. Compound is still the clean baseline. COMP governs protocol changes through Governor Bravo plus a Timelock, with proposals created by addresses delegated more than 25,000 COMP, a quorum of 400,000 votes, a 2-day review period, a 3-day vote, and a further 2-day timelock before execution.

That benchmark matters because it separates token utility from token theater. A token that only signals sentiment, while a multisig or foundation keeps broad spending discretion, does not give holders much economic or governance utility. By contrast, Curve ties veCRV directly to proposal voting and weekly gauge-weight votes that determine CRV emissions, while Maker ties MKR governance to protocol risk management and even recapitalization mechanics through surplus and debt auctions.

The practical rule is simple. If a DAO wants a governance token to matter, the token must sit at the control point of assets, permissions, or recurring economic flows. If it does not, the market may still price the token, but the protocol has not created durable utility. The examples below show that “utility” is not one thing. It is a bundle of explicit rights, and each right needs its own governance design.

Start with a rights map, not a token allocation chart

Most DAO governance failures start with a category error. Teams design the token distribution before they define which decisions should actually be governed by token holders. The correct order is the reverse. First map the rights. Then decide which rights belong to liquid token voting, locked voting, elected councils, or narrowly scoped emergency bodies. ENS makes this split explicit: major matters are voted on by governance token holders and delegates, while elected Stewards handle day-to-day matters including grants, sponsorships, and working group decisions for a one-year term.

Optimism goes further and separates stakeholder classes. OP token holders vote on protocol upgrades, token allocations, inflation adjustments, director removal, dissolutions, elections, and rights protections. The Citizens’ House uses a distinct 1 member, 1 vote model and has powers over protocol-upgrade vetoes, budgets, missions, and the election of representatives to the Developer Advisory Board. That is a stronger design than asking one liquid token to express every interest in the system.

A useful rights map usually has four layers.

• Constitutional rights: upgrade authority, core parameters, dissolution, director or council removal.

• Budget rights: annual spending envelopes, treasury diversification, grant pools, buyback policies, reserve targets.

• Operational rights: selecting service providers, approving RFPs, paying contributors against milestones.

• Emergency rights: narrowly scoped interventions for time-sensitive failures, with ex post review and revocation powers.

Protocols that separate these layers generally produce clearer token utility. Protocols that collapse them into one liquid vote usually end up with either governance overload or hidden centralization. That is not a theoretical concern. ENS explicitly distinguishes executable proposals, social proposals, constitutional amendments, working-group funds, and RFP budgets, and Optimism explicitly distributes authority across tokenholders, chains, apps, and end-users.

Use commitment-weighted voting where capital is long-lived

Commitment-weighted governance works best when the DAO is allocating recurring economic value over time. Curve is the clearest live example. Users lock CRV for between 1 week and 4 years to receive non-transferable veCRV, with voting power increasing with lock duration and then decaying linearly as unlock approaches. Lockers receive governance rights, gauge-weight voting power over emissions, boosted rewards, and a share of protocol fees.

The reason that structure works is mechanical, not rhetorical. If governance decides where emissions go every week, and those emissions affect long-term market structure, then voting power should belong disproportionately to holders willing to lock capital for long periods. Curve’s design makes governance costly to rent and harder to swing with purely short-term liquidity. The token is not just a ballot. It is a time commitment.

That does not mean every DAO should force lockups. Liquid delegation is often better for low-frequency constitutional decisions, because it reduces participation friction and keeps the governance surface legible. Compound and ENS both rely on delegation rather than lock-based voting. ENS holders cannot vote or create proposals until they delegate, proposal creation requires 100,000 ENS delegated to the proposer, executable proposals have a 1% quorum, voting lasts 7 days, and execution is delayed by a 2-day timelock.

The design implication is straightforward. Use liquid delegation for legitimacy and accessibility. Use lock-based or escrowed voting when governance is repeatedly allocating emissions, fee streams, or other long-duration economic rents. Mixing those cases together under one unconstrained token usually creates either low turnout or rent-seeking.

Treasury utility is budget authority under hard constraints

Treasury utility is not “the community can spend the treasury someday.” Treasury utility is the ability to authorize specific budget flows under explicit constraints and delay mechanisms. OpenZeppelin’s governance model makes the basic architecture clear: proposals are executable sequences of actions, treasury transfers can be encoded directly in proposals, successful proposals can be queued through a timelock, and the assets and permissions should sit on the TimelockController rather than the Governor itself. OpenZeppelin also warns that giving additional proposer rights to parties besides the governor is “very risky” because they can gain powerful governance pathways or block approved proposals.

That architecture has a direct treasury lesson. Large discretionary reserves sitting behind a foundation wallet or broad multisig are not a governance feature. They are an agency problem unless the discretion is sharply bounded. Optimism’s public documentation is notable here. The Foundation’s allocated-budget treasury requires governance approval to access via annual Foundation budget proposals, and the treasury is described as subject to oversight by tokenholders, chains, apps, and users through annual budgets.

ENS shows another workable pattern. RFPs must specify scope, deliverables, timeline, the approving party, and a maximum budget. Stewards then decide whether the work can be paid from working-group funds or needs a DAO-wide executable proposal, and compensation is disbursed against milestones. That is much closer to real treasury governance than general-purpose token polling.

From a treasury risk perspective, a governance token becomes more useful when it is attached to a budget charter with hard rails. That is also the core of a treasury strategy for token-based projects:

• an annual operating budget approved onchain,

• category caps for grants, BD, liquidity incentives, and contributor compensation,

• reserve and runway targets before discretionary expansion,

• timelocked execution for non-emergency transfers,

• sunset dates or renewals for councils and service providers,

• and mandatory public reporting against approved budgets.

Without those rails, “community treasury” often means token holders bear dilution and price risk while a smaller operational class retains effective capital allocation power.

Make the token absorb consequences, not just opinion

A governance token has stronger utility when holders internalize the consequences of poor decisions. Maker is the clearest example. MKR governs risk parameters across the protocol. When the system accumulates enough surplus Dai above the governance-set surplus buffer, that surplus can be auctioned for MKR and the MKR is burned. When the system has debt and insufficient buffer, debt auctions mint MKR and sell it for Dai to recapitalize the protocol. Maker’s own docs explicitly frame the surplus buffer and debt auctions as part of the protocol balance sheet.

That linkage matters because it turns governance from abstract preference expression into balance-sheet management. Good risk governance can support surplus capture and reduce MKR supply. Bad risk governance can force dilution. Inference: that is one of the few forms of governance-token utility that directly aligns long-term treasury discipline with token-holder incentives.

Most DAOs should still be cautious about copying the cash-flow side of that model too early. If recurring revenue is thin, volatile, or politically needed for ecosystem reinvestment, then promising holders buybacks and burns, fee-sharing, or perpetual treasury extraction can weaken the protocol’s real operating capacity. Optimism’s docs are useful here because they tie treasury management to a capital-allocation model and recurring revenue contributions from OP Chains, rather than pretending that every token needs immediate holder distributions. OP Chains contribute the greater of 15% of net transaction fee profit or 2.5% of gross transaction fees to Optimism, and OP Mainnet contributes 100% of its revenue to the shared treasury.

The treasury sequencing rule is therefore conservative. First establish recurring inflows, reserve policy, and budget process. Then decide whether the token should govern capital allocation only, absorb downside like a backstop asset, or also participate in upside distribution. Doing these in the reverse order is how DAOs turn their treasuries into short-term extraction pools.

Participation design is utility design

A governance token is not genuinely useful if only a tiny minority can use it effectively. Chainalysis found that across several major DAOs, less than 1% of holders had 90% of the voting power. Academic work on Compound and Uniswap found that as few as 3 to 5 voters were sufficient to sway the majority of proposals, while another large-scale study found that contributors held the necessary majority on average in at least 7.54% of DAOs, singularly decided at least one proposal in 20.41% of DAOs, and observed token-ownership shifts before 14.81% of 8,116 evaluated proposals.

The response is not to abandon token voting. The response is to reduce the gap between nominal and effective utility. ENS does this by making delegation mandatory for participation and by publishing clear thresholds and timelock mechanics. Optimism goes further with explicit conduct expectations for active delegates, including a social standard of participation in 70% of votes, conflict-of-interest disclosure, and no self-dealing rules.

A defensible reference design for new DAO governance tokens looks like this:

• Keep a liquid delegatable token for major constitutional votes and representative elections.

• Use a locked or escrowed voting layer only for emissions, fee-allocation, or other repeated long-duration capital decisions.

• Put treasury assets and privileged permissions behind an onchain governor plus timelock, not an open-ended operational wallet.

• Push day-to-day execution to elected councils or stewards with fixed mandates, fixed budgets, and fixed terms.

• Keep emergency powers narrow, removable, and auditable.

• Write a treasury charter before launch that defines reserve targets, diversification policy, budget cadence, and what requires token-holder approval.

That is where token economy design stops being cosmetic and starts becoming institutional design. In FinDaS Tokenomics’ work, the highest-leverage deliverable in tokenomics consulting is usually not the emissions spreadsheet. It is the permissions matrix, budget charter, and reserve policy that determine whether a governance token actually governs anything.

For most DAOs, the winning structure is not “one token, one vote, over everything.” It is a narrower and more disciplined stack: liquid delegation for constitutional legitimacy, time-commitment for rent allocation, elected operators for day-to-day execution, and treasury permissions that are constrained before they are decentralized. That is how a governance token stops being symbolic and starts functioning as real utility.