How to assess a DAO?

The valuation of new organisational models is always hard. Clear-cut definitions are not yet established, benchmarks are missing, and time-proven best practices of analysts do simply not exist.

Since the ICO boom which peaked in late 2017, we have learned to use certain metrics that can help when assessing the value of tokenised projects.

To assess infrastructure projects (e.g. new blockchain protocols like Solana or Avalanche), one can look at on-chain metrics such as the number of new addresses created or the number and size of transactions.

For Decentralised Finance (DeFi) projects, of which most have not existed before 2020, Total Value Locked (TVL) has been established as a proxy of trust — the more money users (and other projects) put into a project, the higher its success is perceived.

But how to evaluate Decentralised Autonomous Organisations (DAOs)?

Definitions first: What is a “DAO”?

A multitude of definitions exists on what constitutes a DAO. Some refer to very minimalistic forms like “a telegram channel and a splitwise account”. Thus, for the average reader, it’s hard to draw a line to distinguish if a project that calls itself “a DAO” is in fact one or not.

While the term DAO emphasises it, most DAOs are neither particularly decentralised, nor fully autonomous, nor really organisations.

Decentralisation can somewhat be justified by the fact that some functions of a DAO, like a particular ERC-20 token contract it is using, are built on a decentralised blockchain protocol and usually no central headquarter in a physical sense exists.

Most organisations are most certainly not fully autonomous, as human actors who execute the proposals that were voted for are still necessary. Some DAOs attain a certain level of automation by using oracles (such as zodiac by Gnosis Guild) or other software tools that facilitates their voting process. Yet, in a strict sense, most DAOs should call themselves “Decentralised Communities” instead.

An organisation is defined as a group of people who work together with clearly defined rules, roles and responsibilities. In many DAOs, access criteria are simply defined as the ownership of a certain token. While some DAOs are well-functioning organisations, in some cases, token holders do not even realise that they are considered members of a DAO.

Given the many shortcomings of the term “DAO”, we propose a minimum definition of a

DAO:

— operates on a sufficiently decentralised blockchain protocol \ — includes a secure shared on-chain account to store value in the form of tokens \
— any transaction that involve the DAO’s assets are to be permitted by clearly defined rules \
— takes governance decisions together in a transparent way (like a “trusted notary”) \
— clearly defined criteria for membership, such ownership of a specific token

Metrics

DAOs that meets this definition can be extremely heterogeneous as we shall see shortly. To assess a DAO and compare it to others, no single figure will suffice. A mix of quantitative and qualitative metrics can be combined using a scoring model. The metrics below are a first attempt to point out areas to look at closely but are in no way complete.

Strategic metrics

Purpose

Any DAO should have a clearly defined purpose that is communicated precisely. Whether it is to invest in specific assets or to improve the world by funding ecologic initiatives, the most important thing is to have a common goal. Ideally this is expressed in a concise vision and mission.

As the management expert Peter Drucker famously claimed, “structure follows strategy”. Thus, the form of a DAO should follow the strategic goal of the DAO. While some DAOs are ephemeral and used for short-term endeavours like bidding for a copy of the US constitution, others are built to endure. The metrics below are mostly applicable to longer-term oriented DAOs.

Product/Service

Classical organisations can provide any product or service. Analogous, the product or service a DAO provides can be really anything, too. It can be a blockchain protocol, an off-chain resource, an investment fund or a specific service (Binance Research, 2019).

Resistance to competition

As for any human endeavour, others might have had the same idea before. Enough business literature has been written about the topic of competition, so there is no need to go into detail here. Given the open-source nature of DAOs, one additional risk to be considered is the possibility to “fork” an existing DAO and compete directly. Good DAOs have clear moats preventing this working for them, such as high loyalty of their members, economic incentives, high switching costs and a strong brand.

Age

A DAO that has been in existence for multiple years, has seen bear markets come and go and still thrives is obviously stronger than a new DAO that came into existence only a few weeks ago. To interpret some of the metrics below, age can be taken as a denominator to see achievements normalised by time.

Legal structure

If a DAO incorporates a legal structure, and if so, in which jurisdiction, is a strategic decision. The reason why many protocols launch a DAO and hand over power to them is often to avoid regulatory exposure of the founders personally.

In contrast, some DAOs incorporate an additional legal entity to interact with the traditional financial industry, to make payments in fiat currencies and to pay for developer tooling of engineers. For some forms of DAOs, such as for-profit investment DAOs, a well-recognised legal structure is recommended while for others, Ricardian Contracts can also be a potential solution, as argued in @thelaoofficial/a-taxonomy-for-laos-making-sense-of-the-emerging-lao-ecosystem-1122b035fe1a">a blog post.

Financial metrics

Size of DAO treasury

To simply calculate the current USD value of all assets of the DAO locked in smart contracts (such as a multi-sig account) is usually straightforward. However, there might be more than just one account. Let’s take Gnosis as example: They own a ton of ETH from their ICO in 2017, so the team is not going to run out of money anytime soon. Yet, there is no direct connection of the GNO tokens locked in the DAO treasury and the ETH owned by the team.

A good place to compare DAOs ranked by their treasury sizes in USD is DeepDAO.io.

Treasury diversification

The trend of the DAO treasury size in USD is also crucial, but be aware that the USD value fluctuates with market volatility. To minimise the risk arising from these market movements, diversification is important for any investor and DAO treasuries are no exception. Many DAOs hold almost no assets except a large percentage of their own governance tokens. This can lead to a downward spiral in price if the DAO sells a large junk at once.

Some DAOs, such as PrimeDAO among others, have started to use token swaps and traded their governance tokens with tokens of other DAOs. While this is a good start, holding various tokens of a narrow space (e.g. “metaverse” projects) still exposes the DAO to sector risk.

Ideally, a DAO holds a broad mix of growth assets, fixed income bearing assets (e.g. yield bearing DeFi tokens) and tokenised real-world assets (e.g. tokenised real estate).

Depending on the nature of a DAO and the assets it is holding, individual KPIs should be defined. A positive example is MakerDAO, who uses KPIs tailored to their situation such as “Interest Income Correlation” among others.

Revenue streams

Like any organisation, a DAO needs financial capital. To be sure, income can be gained in the simple way of selling membership tokens with governance rights to new members. Ideally, it’s more than just that. Owning rights to other sources of revenue, such as a DeFi protocol, that generates income which automatically flows to the DAO’s treasury is the superior way.

Burn rate

A classic metric well known in startup valuation; the burn rate measures the monthly expenses of an organisation. The runway it has is thus defined as the time left before running out of cash. Given the high volatility of many tokens, this should be treated with extra caution and wide margins of safety.

It can be hard to assess the future burn rate in USD, bit this is not always necessary. If all expenses of a DAO, such as grants and salaries are made in a certain cryptocurrency that the DAO holds in its treasury, short term fluctuations of the USD price matter a lot less.

Accounting

To quickly assess financial health of a DAO, transparent accounting is very helpful. While the open nature of public blockchains enables anybody to inspect the transactions going in and out of the treasury, additional dashboards and reports can be helpful to save time. A good example is the accounting dashboard by DAOhaus that shows on-chain metrics in a neat layout.

Technical metrics

Core architecture

Most DAOs are based on templates such as Aragorn, Colony or DAOhaus (El Faqir et al., 2020). These parameterised platform DAOs allow anyone to launch a DAO without Solidity coding skills. Just some parameters need to be supplied. This is not necessarily a bad thing, as well-tested code is safer than custom written smart contracts. The other class of DAOs are self-coded — here it is much harder to assess the features and safety. If a self-coded DAO has no significant capabilities exceeding those of template-based ones, it should probably be treated with care. Review of proper testing and the presence of one or more security audits are always advisable, but for sake of brevity we won’t dive into detail here.

Additional tools

In addition to the smart contract-based core of the DAO, a wide range of extensions exists in the form of various tools. A blog post by 1kx provides a great overview.

Voting and execution

Voting can be facilitated ranging from simple polls in telegram or discord to direct interaction with smart contracts. While on-chain voting is clearly more secure and offers censorship resistance, off-chain voting is much cheaper. Researchers found, that members participate less in governance activities if they have to pay high gas fees (Faqir-Rhazoui et al., 2021). Thus, it can make sense to trade a bit of security to gain participation of voters.

Snapshot is the leading off-chain governance tool that allows anyone to create a space for managing proposals and define strategies for creating proposals and voting. These proposals can be voted for by signing messages using a 256-bit ECC private key. The address corresponding to this private key needs to own certain assets (such as specific ERC-20 or ERC-721 tokens) to prove its membership to a DAO.

The crucial link between these off-chain votes and trustless on-chain execution can be established if the DAO is using a tool like SafeSnap, which connects snapshot with Gnosis Safe, the most widely used multi-sig. Generally, beware of DAOs that manually execute transactions, even if they follow the outcome of a vote! While most proposals (such as standardised grants applications) can be handled automatically, some proposals can only be executed manually (such as complex technical upgrades).

Governance metrics

To provide a benchmark we obtained relevant data by querying the public GraphQL API of snapshot, the most popular tool to facilitate signature-based voting. We stored the query results in a relational database and used serval SQL queries to assess the data. In total, we analysed the data of 5000 spaces on snapshot, including 35k proposals and 518k votes.

Number of proposals

A DAO without any governance actions is just a passive vehicle. Innovative organisations let their token holders participate and in healthy organisations, they do speak their minds. Thus, a high number of proposals is indicating functioning governance, albeit with the caveat that the proposals originate from multiple actors and are of high quality.

Our analysis of snapshot data indicates that the average number of proposals by space is just nine, while the median number of proposals for all listed spaces is a mere two.

Proposal threshold

The bar to issue a proposal is different, while some DAOs set it at a minimum of one token, others have much higher requirements that de facto exclude anyone else than original team members or large investors.

Our analysis of snapshot data reveals that about 50% of all DAOs use ERC-20 tokens as a membership criterion, while 13% check for ownership of NFTs (12% use ERC-721 while 1% use ERC-1155).

Proposal acceptance

While a high percentage of accepted proposals looks good on first glance, it could hide other issues, such as who decided the outcome.

On snapshot, 86,3% of all votes are “Yes” votes. Please note that this number does not reveal anything about the weight of a specific vote.

Vote participation

A high participation in terms of tokens can be misleading. An important question to ask is if it is just the core team voting on their own proposals or others are participating as well. https://deepdao.io shows this metric for most DAOs that are listed.

Our analysis of snapshot data indicates that while the average number of votes per proposal is 92, the median number of votes is just 6, indicating that only very few DAOs have a lot of active members that vote for proposals. Furthermore, increasing participation (best measured as the number of individual addresses) is obviously a healthy sign.

Token distribution

An assessment of the distribution of tokens among its holders is important for any project with a token. For ERC-20 tokens a distribution chart is easily accessible on Etherscan for example. It also makes sense to look at it for DAO governance tokens. One metric that can be used to assess the token distribution at one glance is the Gini coefficient. A helpful reference on the subject is the paper “Decentralized Finance, Centralized Ownership?“ (Nadler & Schär, 2020). For ERC-20 contracts, the publicly accessibly website https://bloxy.info shows the metric, although it is advisable to not blindly trust the numbers. As a majority of governance tokens is often kept in DAO treasuries, these tokens should be ignored for the calculation. The Gini coefficient should not be overused, however.

Community

Number of members

The total size of the community can be defined as all members. Who exactly a member is, differs case by case. Some DAOs count any member of a discord or telegram group a member, while others only view holders of specific tokens as real members. Research shows that a DAO with less than 20 token holders can generally be ignored (Rikken et al., 2021).

Commitment of contributors

Most members in any online community are lurkers who do not contribute meaningfully but are rather passive according to research (Nielsen, 2006). This metric is hard to assess and likely involves manual screening of chat rooms, forums, reddit or other platforms for levels of interaction. A good sign is usually if there are several active moderators around who are not a part of the core team.

Net Promoter Score (NPS) could be a good tool to frequently assess the mood of members and contributors.

Social media outreach

A good proxy would be the number of Twitter followers and the size of a subreddit among others.

A good example of a strong DAO community is MakerDAO with an active Reddit community of about 33k members and about 195k Twitter followers.

Concrete example: PrimeDAO (as of 01–2022)

Having listed several useful metrics to assess a DAO, let’s get practical. In the following table, we present a simplified assessment of PrimeDAO using the metrics from above:

PrimeDAO assessment

Conclusion

One could always go deeper whether it is in the area of HR (e.g. analyzing each of the core contributors in detail), Technology (e.g. analyzing each technical capability in detail) or Marketing (e.g. study the effectiveness of DAOs memes). In future posts, we will provide further insights into the rabbit hole of DAOs.

We still hope our indicative guide of assessing a DAO using several metrics was insightful and sharpens your own critical view on DAOs.

Disclaimer：

1. This article is reprinted from [medium], All copyrights belong to the original author [Ferdinand Regner]. If there are objections to this reprint, please contact the GateLearn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. Translations of the article into other languages are done by the Gate Learn team. Unless mentioned, copying, distributing, or plagiarizing the translated articles is prohibited.