The Future of Governance is Modular

Onchain governance has failed to empower organizations to reach their goals. There isn’t a credible person in the industry who disagrees. 

The difference now is that it’s indisputable. The biggest and most well-funded DAOs in the ecosystem are struggling. These projects are unable to adapt over time, and even after spending millions of dollars attempting to do so, end up putting their users at risk while stagnating slowly to death.

The good thing is, we can finally move on and we have a path forward. Even more, we have no choice—governance isn’t optional. 

At Aragon, we’ve experienced this firsthand. Our own challenges, alongside those of Compound and Arbitrum, amplify what we’ve learned and validated all to be true. Monolithic tools that are widely used today need to be replaced with modular ones. They know, we know, and the industry is now starting to wake up to this urgent reality. 

We’ve spent the last 3 years consulting with leading industry projects like Lido, Polygon, Mode, Curve, and many more, to begin fixing that. New tooling, at the very least, must allow for: 

1. The best decisions to be made by the best people: allowing organizations to tailor decision-making processes to their unique goals, risks, communities, stakeholders, and internal cultures—overcoming one-size-fits-all governance models that slow down decision-making and put projects at risk. 
2. Future-proof and adaptable organizations: A modular framework enables organizations to more safely and affordably adapt over time, evolving their governance structures as their needs change—resolving the problem of DAOs becoming entrenched in their existing tools, struggling to upgrade them over time. Every organization must adapt over time, or die.
3. Network effects that benefit everyone: By allowing for the reusing of existing components, a modular framework reduces the need for custom development, allows a project’s resources to be redirected back toward the thing they do best—delivering on their own value proposition—and reduces the security risk surface of newly introduced code. No more forking and re-writing, just installing. 

One-size-fits-all governance doesn’t work 

Organizations are groups of people organized around a shared purpose. Governance is needed when there is division in opinions, beliefs, or preferences held by its members about actions to take in realizing this purpose. How well they resolve these divisions are ultimately what determines their success. 

Whether it’s a web2 startup, a non-profit foundation, or a parliamentary democracy, we know different kinds of organizations need different kinds of governance. 

The next generation of organizations are moving from enforcement of their governance by legal frameworks to automatic enforcement codified into smart contracts. However, this new software is significantly less mature than the legal frameworks of their traditional counterparts. These new orgs are still facing the same tangled mess of human (and non-human) actors with their own diverse interests. 

Onchain governance is enabled by the capabilities of smart contracts, and not the other way around. When the industry moved toward a more representational form of governance, it’s not surprising that vanilla delegate-based token voting became the de facto implementation; it was simply easier to extend ERC20 governance tokens to have a delegate function than build a complex system from the ground up.

That doesn’t mean that all organizations will benefit from this implementation. To the contrary, these existing governance frameworks lack the flexibility to account for the complex internal realities of each project. Just as one-size-fits-all governance doesn’t work for traditional organizations, it won’t work for onchain organizations.

Modular systems are built to embrace the diversity of organizational forms. They are even more important in a maturing industry where change happens fast and often, making the ability to securely adapt imperative.

The hidden cost of monoliths

The leading frameworks for onchain governance are modular from the perspective of smart contract developers who copy paste code from Governor or fork their favorite projects. They aren’t, however, modular for this second generation of DAOs using them, who ultimately end up bearing the costs and risks of their inflexibility.

Customizing these frameworks requires projects to hire developers to modify the contracts and pay for audits. Millions of dollars can be spent annually for a single DAO’s maintenance—going to security and custom tooling services. This diverts resources and attention away from a project's core development activities. 

For pragmatic founders leading some of the most promising new projects in the industry, these high costs make governance an afterthought—they often don’t even know where to start. It’s an especially difficult first step because they know they could be stuck with their choices.

Smart contracts are "sticky" with high migration costs. But, unlike the smart contracts they’re being built on, organizations aren’t immutable.

As an organization matures and its market evolves, the complexity of stakeholder groups and internal functions grows. Their governance needs change. 

The ongoing costs of maintaining one-off, customized monoliths accumulate over time. These costs do not accumulate on the tooling provider; they accumulate directly on the project to the benefit of the tooling provider. Maintenance effectively turns into a money pit.

Even worse is the added security risk introduced by code changes. This is the exact reason why Offchain Labs recently halted the myriad of proposed upgrades in the Arbitrum DAO.

Projects wouldn't have to take on the full burden themselves if they could leverage the network effects of shared interoperable components built by an evergrowing community of developers. They would be more readily able to tap into these network effects if the tools they’ve chosen were built with this in mind from the start.

Aragon OSx: permissions as a primitive

The marketing team doesn't usually need to be an admin on your GitHub, just like the CFO doesn't need the Twitter password or banks shouldn't (in theory) be allowed to freeze your assets. 

It’s no different onchain. Every action—defined by a solidity function signature—can be guarded according to the rules of an underlying permission management system. 

Most onchain governance frameworks have some system for handling access control - it’s just a matter of how flexible and interoperable they are with other primitives. For example, Governor’s governance primitive is delegated token voting, but developers can extend it by adding an access control system directly to the Governor. Safe’s governance primitive is a multisig, but developers can add modules that have their own role system. 

Aragon OSx flipped this on its head by making permissions themselves its core primitive. This allows us to put an organization and its assets at the center, and push the governance to the edges. An organization may in the future no longer need token voting at all, but they will always need permissions.

For this reason, we expect a gradual convergence towards a common permission management standard, effectively making Aragon OSx the hyperstructure for onchain governance.

Plugins: making “governance legos” a reality

The flexibility of OSx permissions is derived from their simplicity and standardization. They delimit who can execute actions. 

Permissions are granted or revoked to any Ethereum address, not simply “people” controlling externally owned accounts (EOAs). Namely, permissions can be granted to an OSx plugin address. Plugins contain governance logic inside of reusable contracts. Each plugin is limited in scope, for example in a token voting plugin or multisig plugin. 

Plugins are built to be atomic, with single-purpose logic that is easier to reason about and compose with one another, rather than having to detangle it from a large monolithic architecture. This allows organizations to combine multiple smaller interoperable pieces, mixing and matching them to compose entirely new governance designs uniquely suited to their needs and adapt as their project evolves.

Plugins can be easily deployed by our factory contracts and installed without writing a single line of code—organizations simply need to grant the right permissions directly to the plugin. This approach reduces the risk of error and security vulnerabilities inherent in custom implementations. If a new plugin ends up needing to be built, they will only have to audit the code that is necessary. 

Modular governance patterns

Modularity is the antidote to the limitations of one-size-fits-all governance. When the pieces of modular systems come together, the whole becomes greater than the sum of its parts, unlocking a new realm of solutions.

Our approach to modularity is informed by conversations with countless projects and learning from their needs. Even though every project needs something unique, there are always patterns.

Flexible governance processes 

Governance processes are inherently complex, involving multiple stages that should work together more seamlessly. In the current state of onchain governance, these workflows often depend on disconnected tools like Discourse for discussions, Snapshot for offchain signalling, and an onchain voting mechanism for execution. These fragmented handoffs are inefficient and error-prone. 

To address these challenges, our modular framework is based on three constructs - Processes, Stages, and Bodies:

• Processes represent distinct governance activities, or “proposal types”, required to execute specific tasks or "actions." A DAO can have any number of processes.
• Stages are the sequential phases within a process that a proposal must progress through to be approved and executed. A process can have any number of stages.
• Bodies are plugins that evaluate and update the status of stages based on their governance logic. A stage can have any number of bodies.

By codifying Processes, Stages, and Bodies into a flexible, audited plugin, we have generalized the need for governance to take place over time, supporting any extension of this pattern, without requiring custom development or auditing. This gets us flexible and customized governance setups with all the benefits of being onchain but in a more secure, efficient, and scalable way.

Optimistic governance 

Optimistic governance allows proposals to pass by default unless actively vetoed. First pioneered by Aragon with Govern and Court in 2021, it is now a native feature of any governance process built on OSx. Any body can be configured as a vetoer at any stage of any governance process.

By “optimistically” assuming success unless vetoed, optimistic governance significantly reduces governance overhead. Stakeholders don’t need to participate in every decision. Routine or non-controversial proposals can pass with minimal oversight, freeing token holders, busy board members, or unrelated teams from constant participation. This approach prevents voter fatigue while still giving oversight when necessary. 

Unlike the more common all-or-nothing token voting of monolithic frameworks, which burdens voters with every decision, optimistic governance lets organizations stay focused and safely ship.

Proposal types 

Not every decision—financial, technical, or strategic—should follow the same process. Modular governance allows each to be handled separately, with processes granted only the onchain permissions they need, eliminating reliance on trusted intermediaries and offchain bureaucracy to manage the different decisions an organization needs to make.

By granting different permissions to different processes, a single DAO can achieve true "separation of powers," where different bodies are limited to participating in the processes they are responsible for and no other body can override them (unless they are specifically given permission to!). For example, token holders might be able to vote to add or remove members of a council that governs via a multisig.

This separation of powers introduces checks and balances, addressing the flaws of monolithic, token-based systems like Governor, where a super-admin acts as god.

It’s also more efficient. For instance, zkSync’s deployment of three separate DAOs for different governance processes triples their maintenance costs and complexity. With proposal types, these processes could be consolidated into just a single DAO deployment.

Adaptable governance 

The pace of our industry is notoriously relentless. Regulatory uncertainty, technological advancements, and evolving stakeholder expectations are constants—not exceptions. To be successful, onchain organizations must adapt.

Modular governance provides the tools for this adaptability by breaking systems into discrete, interchangeable components. Organizations can install, uninstall, and update individual plugins without overhauling their entire DAO. Proven components can often be reused.

This approach builds on three key principles:

• Encapsulation: Modules handle distinct functions, preventing changes in one from disrupting others.
• Interoperability: Standardized interfaces ensure seamless integration of new components.
• Reusability: Components gain trust and over time, accrue “lindy” as they are repeatedly used in different contexts.

Aragon DAOs, unless configured otherwise, are autonomous. This ensures power remains fully within the organization itself, even during upgrades, without reliance on external development teams. Compare this to monolithic systems like Governor. For instance, a recent Arbitrum DAO proposal to upgrade contracts came to a halt amidst a debate around the fact that it wasn’t an upgrade at all, but a deployment of completely new DAOs. This isn’t a new or rare phenomenon and has happened to several large DAOs recently. These debates and the ensuing risks will only become more common as DAOs mature and require change.  

A modular design could have mitigated these challenges, reducing costs and risks associated with the changes. Adaptable governance, powered by modularity, will be what separates organizations that are surviving today from those that will be thriving tomorrow.

Conclusion: The modular governance imperative

Monolithic governance frameworks are putting at risk the very things they are intended to govern – revolutionary DeFi protocols, public goods funding protocols, Ethereum scaling solutions, and all the other exciting things that we are in this industry for.

Despite the clear benefits of modular governance, many projects remain tethered to monolithic tools and extractive providers. Why? The issue isn’t talent—governance professionals and DAO builders are eager to innovate. Instead, they’re constrained by laborious decision-making processes and rigid systems that aren’t appropriate for the actual problems they’re trying to solve. It’s a difficult loop to innovate your way out of. Delegates with no skin in the game and tooling providers have incentives to reinforce the status quo of unchecked vanilla delegate voting, yet are themselves the decision makers.

But things are changing. Governance experts and communities are recognizing the need for modular systems that unlock them to do better work and set their organizations up for success.

Aragon OSx is the framework to make this shift, fully onchain and trust-minimized. With flexible governance patterns like Processes, Stages, and Bodies; optimistic governance at any Stage to reduce overhead; permissioned proposal types for true separation of powers; and adaptable frameworks that scale with organizational needs, OSx organizations are scalable, efficient, and future-proof in a rapidly changing industry.

Since 2017, Aragon’s mission has been to prove that onchain governance can help organizations reach their goal. The same conclusion is reached, again and again: the future of governance is modular. 

Stay tuned as we have some huge updates coming in the next few quarters that will further unleash the power of modular governance.