What actually motivates DAO delegates?

DeXe Protocol
5 min readSep 8


One of the major differences between DAOs and other types of crypto projects where people can buy shares in the form of tokens is that DAO tokens give holders voting power in the DAO, which, of course, is why they’re called governance tokens. But are they actually being used for voting?

Apparently, the current answer is a lot closer to No than to Yes. The Dune Dashboard chart of Nouns DAO voting below is fairly typical: unique voter turnout hovers around 7–20% of all tokens held, often dipping much lower.

If you take away the voting power of the core team, early investors, and whales — how many actually unique voters are participating in governing?

The other question is the competence of the voters. How many voters have a comprehensive understanding of the issues they’re voting on, especially if it’s a very technical or a very particular issue that requires specific expertise? How many even consider the nuances of the issue? Looking at the alarmingly large number of near-unanimous vote decisions (we wrote about them here), one has to wonder.

Luckily, there is a solution to lackluster voting that many DAOs have already implemented: delegation. Depending on the particular DAO, members can delegate their voting rights to any wallet at all, any other governance token holder of that DAO, or registered delegates of that particular DAO.

In theory, this is great: people passionate about a specific DAO present their case for being a delegate and get popular support to make their mark on the DAO with the votes delegated to them. Unfortunately, in practice, delegation is coming up with the same shortfalls as voting in general.

From the slides of Karma’s founder, Mahesh Murthy, you can see even among the top DAOs, delegates are not active in their governance.

It appears that over 90% of delegates and about as much of the voting power of the dYdX DAO has never voted — not once!

Overall, ~55% of delegates across DAOs have never cast a vote, with some DAOs showing much worse results:

And when delegates do vote, look how few are voting against the majority:

This is fine if token holders merely want to cast any vote for any proposal just to qualify for a possible airdrop down the line. But for the actual governance process, this is a disaster.

Clearly, most delegates currently are not especially motivated to vote. So what could motivate delegates to actively — and meaningfully — use the voting power delegated to them?

In a number of DAOs, the top delegates are team members. In others, it may be social media influencers building their own brand. With the former, there is danger of centralization via excessive power concentrated in the team’s hands. With the latter, there is always a risk of the delegates using accumulated voting power to manipulate the DAO to their own needs and not in the interests of the DAO itself, as well as a risk of them having too much power in too many DAOs to pay enough attention to each.

The question remains: how do DAOs attract delegates who do not centralize power and who actively and substantially contribute to the DAO?

Maybe specifically targeting the kinds of delegates the DAO needs is worth consideration. Who is needed? Looking at historical governance systems, the answer seems to be “those who are competent governors.” This means specialists, experts, people with the skills, knowledge, and experience to notice the entire SWOT grid of the DAO (Strengths, Weaknesses, Opportunities, and Threats). Those who can then make knowledgeable proposals based on their expert assessment of the situation. If investing has “smart money” — maybe DAO governance should have “smart votes”?

Naturally, token holders’ lack of motivation should also be addressed. If they do not wish to vote themselves, how could they be motivated to delegate to those who have the expertise and the desire to be effective delegates? Appeals to altruism work — but only so far. At this point, it is shown to be simply insufficient. Many DAOs already implement rewards for voting and delegation — perhaps that can be adjusted to reward active and effective delegation, rather than delegating to those who are really good at Twitter.

What if rewards for token holders were tied to the effectiveness of activity of those to whom these token holders delegated their votes? With delegate reputation dashboards like those of Karma, it’s already easier to research delegates.

Certainly, with the development of better DAO governance platforms, the friction of finding, vetting, and verifying delegates should be lowered, along with re-delegating. Maybe that will lead to a greater percentage of DAO governance token holders being more deliberate with their delegations and to more delegates being more deliberate with their voting.

Ultimately, the goal of delegation is smarter, more active governance. Which incentives do you think could motivate delegates the most? Maybe they’re not monetary at all? Is the delegate selection process flawed? Are there approaches we’re not even looking at (but should)? Share your thoughts.

Stay tuned!

Website | Telegram channel | Binance Feed | Telegram chat | Facebook | Medium| LinkedIn | Twitter | Reddit | Discord



DeXe Protocol

An innovative infrastructure for creating and governing DAOs.