April 1, 2024
What DAOs are good for, and how to push them to the next level.Organizational structures help humans align around common goals. When forming a new organization, it’s important to understand the nature of the task at hand and to choose an organization structure best suited to that task.
Crypto originated the concept of a new kind of organization called a “DAO” – decentralized autonomous organization – which promised to unlock exciting new kinds of coordination on the internet. Over the past several years, DAOs have led to countless experimental projects, lawsuits, and debates. But at this point in time, DAOs have unfortunately become tightly coupled with a form factor that’s holding back the evolution of this idea.
In their most common form today, DAOs involve an on-chain treasury that is governed by token holders through a proposal system via smart contracts. This process is “decentralized” (it’s democratic and on-chain), but it’s also centralized (every decision must go through this single-threaded process). In other words: DAOs today put a centralized decision-making process on decentralized rails.
The problem with DAOs today is that they’re sitting in an awkward middle place between centralized companies and decentralized protocols. They get some things right (broad distribution of token holders, anyone can buy the token), but then turn those things into weaknesses by directing them towards traditional decision-making processes.
I think the main reason this is happening is that 1) we’re overly biased towards a “onchain company” analogy and 2) people creating DAOs don’t have a clear mental model for when they’re the right tool for the job. Here’s a very simple take on that:
Companies are the best tool when you want:
This is achieved through coordination and hierarchy.
DAOs are the best tool when you want:
This is achieved through incentive alignment.
We’ve got companies pretty figured out. But DAOs are suffering from a huge amount of untapped potential. People haven’t really grasped what they’re good for, and their form factor has stopped evolving. We should change this.
The way forward for DAOs is to push their own strengths to the extreme. Set a broad directional goal, then allow people to participate permissionlessly and in parallel (this also probably requires dropping the term “DAO” which now carries strong negative associations for onchain users).
In it’s most basic form, this is just launching a token with a vibe. We see organic parallelized efforts for even the most basic memecoins, where holders participate in content-creation in hopes of pumping their own bags.
In more sophisticated forms, we have a long way to go, even if becoming more sophisticated only means small tweaks at the software level. One good source of directions we could push on is Kei's 2020 blog post on the main qualities of DAOs. If we ’re overly biased towards the analogy of a company today, the correction may be to think about DAOs more like protocols. Instead of thinking about DAOs as people at the center (voting), with software at the edges (tokens), we may want to swing back to thinking about DAOs as software at the center, with people at the edges.
I want to see this happen. Rather than creating an abstract framework for how to do this, I think what's needed is tangible, specific experimentation, starting from a clear link of thinking:
If you have a specific plan, and you want to be precise about execution, start a company. If you have a broad directional goal, and you want to instigate parallelized effort around it, start a DAO. But go all the way.
Get an email when I publish a new essay