Scaling a DAO: challenges and lessons learned - Juan (@0x7u4n) @ETHDubaiConf

Transcript

okay guys let's let's get started uh my name is juan i'm the core unit facilitator at maker dao scs that's sustainable ecosystem scaling and the facilitator what they do in maker is they are the point of contact between governance which is the mca holders or their representation and and the team which has been appointed to do something again by governance this is a bit my journey to the to the davos i worked on different big firms doing consulting some consulting for big clients also then i i was fascinated by startups and scale-ups both in barcelona and berlin

and i ended up in a yeah in the dao world mainly in maker and that's what i'll be discussing today i'll be taking questions at the end so if anyone wants to know about something please let me know so what is a dao i'm not a maxi so i don't want to start a big debate about this is decentralized enough or not there are different opinions about this in general but basically for for those that don't know which shouldn't be a lot of people here but it's a decentralized autonomous organization decentralized usually mean that there's no person to go to there's no legal entity behind the dao and it's governed by a protocol

autonomous means that forms are teams are formed independently they make their decisions independently and they are mandated by the token holders or the the voters and finally the organization part it's uh it's a bit the joint purpose it's the vision the shared community and it means that they need to report to the holder so for me if those if those aspects are there at least present or the intention to get there we're getting some somewhere and again it's not binary it's a spectrum so you can be trying to decentralize even more but what really is a dao it's it's a structure that supports uh decentralized work capital and workforce and these

three pillars are quite important so we're going to be reviewing them quite often but it's basically people the capital to incentivize them to finally do something which is a vision or or work or whatever we want to achieve so at the very early stages when we're bootstrapping a dao usually does look a lot like this right so we have a core team which is being funded by a treasury there might be some vcs outside so so we get we get the funds from there they are developing a protocol hopefully they have some kind of product market fit which means that they have actual users and and they're solving a

real problem this protocol might or might not be controlled up to some aspect by the token holders so they might be developing some kind of governance mechanisms and the community pretty much helps the the core team it's like hey i have an idea or maybe we should do this but it's a core team which is actually developing and delivering on these ideas so yeah scaling it out so what happens once that we move to the next step and i'll tell you a bit what happened in in maker but you might have heard that the maker foundation decided to dissolved last year so a bunch of different core units popped up

which are directly funded by the protocol which is governed by the mkr holders and yeah we do have real revenues you can check them live on chain but the users are the volt users are funding the protocol through the stability fees and then that funding is is pretty much used for the core units to develop the protocol even more and to to grow further there's this little issue with us that you need to be thinking about the final user of the protocol but you also need to think about who your boss is so in a hierarchical

company it's quite easy to understand who your boss is because you see the hierarchy pretty pretty clearly in this case the the mcar holders or any type of token holders it's not particularly clear who they are there are some delegates so that helps a lot because you can at least have some conversations get feedback try to understand their vision or where they want the protocol to get but but that's the that's the main issue that they that they have and the delegates uh on their when they are trying to get something done they need relevant information so they need to have these things ready they need the experts to evaluate the proposal if it's technical they need tools to evaluate if the work

got done or not they need the right incentives to vote so in maker for example you are paid to be a to be a delegate and they need to be able to cut through the noise so we have this which is a nice problem to have i guess but our forum is almost a full-time job just going through it there's a lot of content all the time which is great we love ideas but at some point that that pushes the limits of any any delegate or contributor and the core units on on their on the other hand they need to be well first of all there's a flat hierarchies so this means that no core unit can impose on another current units and this is quite important because if

you want to become a core unit for maker you cannot say i'm going to build i don't know financial structures so that the other core units need to do this because the other core units are first of all already approved and then you cannot be mandated to do something for for the other person so what you have to do if you're a support core unit is a bit tried to to cater to the users lead by example and say hey we built a tool which is really really easy to report on your finances can you please start using it and then it's a bit the delegates or the mcat holders that are going to be pushing for for these tools to be used uh so it's not like there's a boss that comes directly and says hey everyone needs to

do this now unless it's the the delegates and they have the voting power to to vote a core unit out um yeah they can coordinate with other court units it's important it happens a little bit uh in a generic way it's uh like in every other dow if someone shows up and they want to do something usually there's some momentum and and coordination happens almost organically but it uh it's it's not some something simple at all and finally they can be redundant and this is also important and when when we need to sometimes we criticize organizations for not being efficient in a doubt this

might be a little less important as redundancy might be crucial for for that to survive so this is a bit what a dao at scale looks like on the left we have the the centralized organization the hierarchy as i was talking you have an executive team which could be the c suit people or the board members that are giving like more abstract directions and then you have a bunch of middle management and at the end you have the the execution or the people getting the the actual work so this is a little the yeah a little secret of the of the dials but to replace that middle management which serves to actually control and make sure that things are getting done

you need some core units that are focused on transparency and quality control and this is quite important because when they propose things like that it seems like it's a waste of money so it's a waste of funds why are we doing this but at the end it would be quite crucial to give mkr voters or the delegates a view on what's happening across the dial so they're different frameworks or or different tools that they can build for koreans to either self-report or say they are doing what they're doing another example could be the engineering teams are actually documenting their code so there's a bunch of different things that can be done from the dev perspective to ensure that this middle management layer that disappeared has actually a way of um

of still giving the value providing the value that they that they were doing so to recap a bit uh sustainable ecosystem scaling that's my my core unit and what we do is we're helping make a doubt to remove barriers between decentralized work capital and workforce again these are the three pillars that i was that i was mentioning and and they're quite important and today i'm going to talk about three different challenges related to those three pillars so the first one you probably all suffered it is getting talent so it's getting the people attracting them making sure that there's a good culture how they lower the entry barriers for new teams how do you reduce the opportunity cost the second one it has to do with the

with the capital so what challenge should we tackle next towards where should we grow as a protocol how can i help make her grow what are the related costs so those are other challenges that we've been facing and finally managing the work as i was mentioning how do we make sure that the the work gets done and most importantly how do we coordinate people over the danbot number and for those that don't know up to 150 people we're all supposed to be very natural and coordinated as humans after 150 it becomes much more complicated so we've been trying a couple of things the first one is to set or improve frameworks that we already have the second one is an incubation program

and the third one is uh applied research uh grants and i i mean applied research because they are very very concrete in in nature uh so it's not like some crazy phd uh grant so so that's that's what we're going to explore now so on the frameworks um right before the foundation uh dissolved uh i contributed to the mip framework by giving a bit of feedback and especially to the to the core unit framework which as you have guessed it's uh the vision for a core unit the the funding or the or the budget and finally the facilitator which is the the person

responsible uh but yeah we have i think around 70 mips by now this is not new to maker there's a hundred protocols that that have the same thing uh ethereum python i don't know and uh yeah this is permissionless so anyone could come and propose any map in the forum and and they would receive feedback eventually get voted and it could become the the law right uh our governance process it has two components uh the off-chain part and the on-chain part the the off chain is very very public um

with public i mean open so anyone can come you don't need mkr to participate you can comment you can give feedback you can propose your own and and the on chain is very much dictated by the token holders the mkr holders especially the delegates um so that's a bit what we're we're doing and it's delegated plutocracy which means that if you have more tokens or if a delegate has more tokens it has more they have more voting power um so on the mips framework if you people are building uh something similar to govern your dao things that could have gone better from on this aspect well first of all the

core units we didn't plan much for an off-boarding process so when when yeah when the community wants to off-board the core unit or fire core unit it didn't look very pretty for us i mean firing someone is never pretty but it was quite dramatic very public so maybe thinking about how would that work as well the second one is quite important is better templates so the templates that we did were quite generic and very let's see what happens and how people use them right now we're working on improving them but [Music] but it will be good to direct contributors to to give you the information that you need or that hemp

care holders need to actually judge the the proposals then maybe include performance metrics include a roadmap include a plan so those are things that right now might or might not be there and then the community might give feedback like can you please include this but it's it's not in the template so so that will be my my key takeaway and then we have the the collateral onboarding so this mipset is so um we also use the mip framework to onboard collateral so at the very beginning we didn't think about off-boarding processes and and we also didn't think much about the off-chain assets so that's real-world assets [Music] so those are things that we could have

yeah i mean in hindsight it's quite easy but but we could have worked better on them the second thing that uh well the yeah second thing that we tried is the incubation program and this is quite exciting for me it's uh it was run by scs for the last almost year now and yeah why incubate there are many many reasons so usually contributors don't know how to set up a core unit what they should work on what challenges we're facing uh how they can contribute how maker works so all these things are our challenges and our barriers to entry that you do not necessarily need to impose on people

that are interested in in helping um so yeah that's our governance process as i was explaining it and a rough timeline so basically a timeline without incubation for anyone wanting to form a new core unit for maker would almost mean three months of of the team going potentially without compensation for any of the the work that they are being done to to become a team so that's that's why we decided to to test the incubation program the the goal was to reduce this provide them with education capital and and support and to make sure that they successfully

got on boarded by the mkr holders to as a core unit as an independent court unit and this is quite important it's not our community that decides if they're going on or not it's really the the token holder so so that's uh in in one case for example we we did not manage to incubate them the token holder said that uh that it was not that that's not what they were looking for so lessons learned from the incubation program if you ever decide to run your own for for your tao focus a lot on the facilitator in some cases we started with teams and then leadership is not clear there's no there's not a lot of accountability and it becomes quite chaotic

so make sure that they are almost an entrepreneur and someone very self-starter because they will need to show the value to the mkr holders or well in your case the token holders they will need to recruit people to get help they will need to do a bunch of admin work that they might not like so someone that either started something or they worked in startups might have a good profile for that another thing is to focus a lot on minimum viable operations so when you tell someone how much budget they think you need they will think big and they will expose the budget which is great but the governance might not like that so the token holders might be like

that's a bit too much we don't know these people we don't know if they're capable so try to start thinking on minimum viable operations and then they can scale so you can say we are going to work on this problem when eventually we prove that we're our worth we can scale it up and go even further and the the final one our incubation program was five to eight months more or less and the community feedback we got it quite late almost at the time of the votes so my advice would be try to try to get community feedback before so you can adapt if necessary it doesn't need to be binary at the end yes no but you can

adjust the path a lot more if that's what you're what you're looking for in in a successful incubation program and the last topic that i wanted to talk it's the grants program which is uh it's quite uh yeah i really enjoyed this as well uh we didn't have a lot of of structure but this is a bit how how it worked right so again we have the three pillars the idea the funding that was provided by us and the individual was someone that actually wanted to achieve this and then yeah hopefully the grant was a success so again important things about these specific grant programs it was not based on on bounties so it was not very very

well defined and i think that's uh that was a feature in our case it was very much discretionary so we could decide to whom awarded and and how uh but it had to align with our vision right so if someone said hey i want to build a portal about that if that if that for us was not going to scale maker then we would reject it and and tell them to go get the funding in some other units and finally we are the ones asico unit being evaluated based on the output of the grant so that's why we we had to keep to keep a close eye on the on the program so lessons learned on the grants program

first of all focus on the long term i think they're getting someone good in the ecosystem that's interested it's almost much more important than the scope so the scope can be a little bit blurry in the beginning just make sure that you're betting on the on the right person more than the the right idea you don't need to over define that that takes a lot of time and effort and and that might make things harder the other one is speaker wars the grant is the means it's not the end and as i was saying before uh if if it doesn't align with your vision or what what you are trying to achieve uh it might not be worth pursuing it you're not there to run a grants program

you're there to achieve a vision so so you should focus on that and the last one uh we already mentioned it but it's keep it loose and check in often so ideally even once a week is not uh [Music] if if that's uh if that's good and it doesn't need to be with you it could be someone with your team but having like a regular meeting where you can discuss how things are going they can tell you hey i'm having a hard time finding this information so you can maybe help them so i think that that's something that's quite good in this in this context so yeah those are the learnings i don't know if anyone has any questions thanks um yeah i just had two quick

questions um if you don't mind one was how do you ensure accountability within the dow that these these grants or contributors are actually meeting their goals what methodologies have you actually put in place for that what mechanics and the second is how are you reducing the noise that you mentioned that's trickling down from the community potentially or the token holders to those core units what methodologies do you have to to filter that noise thank you so for the what was the first one first one was accountability oh yeah so accountability i think that that depends a lot on the on the dial

right now in general what i see in different dials is that you have a lot more a lot more funds than than talent so you might take riskier bets meaning that you can allow for someone to scam you or not work for a couple weeks but in the example of the ground if you're checking in every single week and you can see the progress you can see if it's going on track or not and again because we are being judged as as grantors i want i don't want to have a grant that didn't go well uh so i would i would cut that beforehand uh second thing super important is to set the expectations uh right uh so again that's something that maybe

it's clear for for one but it's not and it takes some time to to yeah to get it right and and finally as i was saying in the beginning i think that there are some groups that could focus on developing either tools or or mechanisms for for everyone to report and if everyone is following a best practice and it's quite easy to do and your team is not then the token holders might come and ask you why are you not doing it if it's so easy uh like why are you not being as transparent as we would like to so so i think that's that's on the accountability part and the

the second one was um in relation to the noise that can often come trickling yeah through token holders community um and how you protect or how you you know what you set up for your core units to to succeed that's a i think that's a tougher one or we haven't solved it yet what we do have is some core units uh that are trying to produce summaries uh for example we have a group doing a forum at a glance every week so they would release hey these are the so you can get like almost an email that's like a summary of important things and then you can dive deeper the issue is that you are introducing a bias right like maybe

maybe some issue that was actually important for you it's not it's not there so you're missing it so i don't think we have solved the the noise yet um again i think that the important things need to be need to be tackled by almost by professionals so yeah right now we're trying to get uh people to to join our transparency and quality control track and to be able to produce frameworks to produce reports and to show the vmcare holders this is what's happening with the dow this is an overview of the dao ideally in some dashboard so we do have a couple of uh really good dashboards but but not for for that yet so we're building building that as well

great thank you thank you any more questions thanks oh just um if you're like a centralized type organization kind of like a more traditional one and you want to kind of um or you aspire to kind of transition into a dow type model uh how quickly should that occur and like what are the hallmarks of things going the right way and what are like milestones that are kind of like emblematic of things going the wrong way with regards to transitioning from the centralized organization model to a dow model so one of the things that dao is not is not a corporation on chain i would find it quite hard to

decentralize a company that was born centralized that said i do think that it's possible to be born centralized with the idea of being decentralized at some point so you start preparing quite early and there's a there's a lot of things that you can start doing uh from the tokenomics part uh how are you going to compensate contributors um yeah for me one of the the pillars and again we can go back to the three pillars but you need to have a community that that is going to to take it up when whenever you you leave it running the second one is that you have some kind of product market fit ideally that can sustain that full community and if if you have even more than even better

and and yeah the the last one is the the product which is the vision realized uh ideally you have some kind of uh of draft of what's happening and for me those would be quite important hallmarks to hit before actually turning off the the whole foundation parts otherwise i see it quite complicated to to manage something in a decentralized way thanks [Music] uh in your opinion uh what are the strong signals uh to qualify this is a good dow a good performing doubt to join or like this dao is not doing well and maybe i should enjoy it does that make sense like which token should you buy

i'm just asking on criteria and in general like you know like what's a good doubt to join in in general like how i i would know this this dao is promising and like there is sustainability if that makes sense like you know i know a lot of dows they are not there anymore like one year ago so how can i early know these signals if that makes sense what i personally look for is first if i if i align with the with the vision um and it doesn't need to be save the world or the oceans it can be something like i like the product and i like the way they are approaching it for me the team is super important so i

i will try to have a couple of conversations if they're friendly and if they are smart i think that's another good sign and then another one that i would look at is the sustainability part and making sure that they are getting some kind of revenues or at least that you say okay they they are based from a from foundation still because they haven't found product market fit but they should have a runway for and sometimes that the wallet is unchanged so you can even check uh this is the amount that they're spending so even if they double it it should last for the next two three years so it might be a good bet to to join so i would i would look on on on those things but it it's very personal

and i think that's that's a bit the beauty of it that uh anyone can almost work on the doubt that they want and yeah by the way almost everyone probably knows it but i think that the the price of admission is the is wanting to be there right so when i interview someone is i'm almost trying to show them all the bad things and if at the end of the interview they're still like okay i really want to join then i i know that i have someone but if i need to go out and convince someone i think that the the hurdles and and the yeah the fact that everything's so new it's uh it's quite challenging in general for for anyone so so that makes it harder so i tried almost to dissuade them to to

join even if if i would like to do to do otherwise thank you anything else because i've got questions of my own should yes so i want to know um what are the numbers so how many core units do you have how many uh did you fire and how many did you on board through the process i got that i put my my slide um this is a snapshot from from last quarter um so yeah cs is right there maybe the mapping is not perfect but sometimes it's hard to to group them in in kind of groups so the ones on the bottom right are the ones that

graduated from the incubation program uh the one on the bottom left is the one that was off-boarded and yeah those are a bit the the actual numbers uh so it's the only one that's crossed yeah yes and no more were off-boarded not yet yeah we're working on it i seriously and do you like try to abort people and not teams in general uh no that's that's hard one right because the the dao has an agreement an unchained agreement with with the core units and its facilitator so it's up to the facilitator to

to get to that vision and make sure that they're delivering on what they promised um but we had for example a challenge last when this coordinate was aborted there was one contributor that said i wasn't being i didn't get paid by this core unit they racked me so so it's like okay should the dao be responsible for that or not because the dao doesn't have an agreement with the contributor so it's it's a bit up to the contributor but then you might be losing face uh you might be losing reputation so those are things that you need to balance and again if you're ever building a framework try to to think about that and make it quite explicit because you'll be surprised to the length that people go to

not follow things okay and one more question is regarding the enumeration so how is it structured uh what is the fixed price and what how many like do you pay in stables do you play in your own token how does it work so right now you could potentially so when you're when you're proposing your mipset uh you propose the map for the vision you propose the meat for the budget and the one for the facilitator the one for the budget could be more than one actually and you could actually split it so for example you could say okay i need this meep uh approved and usually it's in dai

which is the operational wallet and this is how much i need to keep my team running then you have the mkr compensation which is a bit like the the token from the protocol i would say most core units have that as well and um so that's a bit like a bonus and you have vesting and and we're following those those things that's another interesting one for frameworks if you guys want to discuss later and what some teams do is that they propose an additional budget so he's like i would also like to do this project if you want me to do it then i would need this extra amount it could be in that it could be in mkr and and if it gets approved then it's a

bit of a green light to to go on and do that okay thank you very much so it seems that we are done thanks a lot for your presentation thank you for anyone who is building in dao it's definitely valuable we are there as well so yeah we have to learn you