# Optimism Retro Funding: So Far, So Good, So What!? by Carl Cervone | Devcon SEA

- Speakers: [Carl Cervone](https://streameth.org/speakers/carl-cervone)
- Channel: [Devcon](https://streameth.org/devcon)
- Date: 2025-10-07
- Duration: 25:43
- Watch: https://streameth.org/watch/yt-pz4vGh53qSo
- YouTube: https://www.youtube.com/watch?v=pz4vGh53qSo

## Description

So far, over 50M OP has been awarded to projects with no strings attached. So good, another 800M OP is planned for future rounds. So what ... is the impact? My talk will offer an objective, data-driven perspective on the "so what" of Optimism's Retro Funding. It will include analysis on how different cohorts of projects have performed longitudinally across a variety of growth and quality metrics, while controlling for different funding and market-related effects.

Speaker(s): Carl Cervone
Skill level: Intermediate
Track: Coordination
Keywords: RPGF, Collective Intelligence, Open Source Software, grants

Follow us: https://twitter.com/efdevcon, https://twitter.com/ethereum, https://warpcast.com/devcon
Learn more about devcon: https://www.devcon.org/
Learn more about ethereum: https://ethereum.org/ 

Visit the https://archive.devcon.org/ to gain access to the entire library of Devcon talks with the ease of filtering, playlists, personalized suggestions, decentralized access on Swarm, IPFS and more.

Devcon is the Ethereum conference for developers, researchers, thinkers, and makers. 
Devcon SEA was held in Bangkok, Thailand on Nov 12 - Nov 15, 2024.
Devcon is organized and presented by the Ethereum Foundation. To find out more, please visit https://ethereum.foundation/

## Transcript

[Music] thanks everyone um before getting into crypto I did work in the coffee supply chain and I've been getting a lot of questions about where the best coffee is in Bangkok and I'm still looking for an answer so if anybody knows where to go for some excellent specialty coffee please hit me up would love to check that out um today I'm going to be talking about optimism retr funding uh it's a experimental grants program being run by the optimism Collective and as a data Enthusiast it creates this incredible substrate for looking at the effectiveness of different Grant programs um I'm the co-founder of an organization called open source Observer we're building a free analytics platform and the community of data scientists who care about measuring impact in the open and so today we're going to be talking about the impact that optimism is achieving um and identifying some opportunities to improve and double down on where that is has been most effective um so I guess just a quick check um how many of the people in this room are familiar with retr funding optimism have all right pretty good so I have a few slides that will be a bit of a cursory introduction to optimism I think the main point that I want to get across here is that I believe this is a really important experiment for the world to pay attention to and to understand regardless of whether it succeeds or fails in the long run there's something special and different that is happening here and it's worth paying attention to so another chart um or first chart of the day we have a lot of charts coming um but it has a another poll how many of you have seen this before how many of you are familiar with the Electric Capital developer report have you seen this all right not as much uh awareness this is one of the U most looked at reports that the crypto industry puts out and they track a metric called monthly active developers um they've been putting this out for for quite some time now it's basically indicator of how many developers there are working across all of crypto not just ethereum every ecosystem is is captured here and so what you can see from this chart is that um things went up a lot in 2021 they kind of plateaued in 2022 and now things have gone down um in aggregate from a peak of around 30,000 to about 26,000 monthly active developers by this metric to put this in comparison I recently heard that uh Gemini which is Google's AI platform they have 1.5 million active developers by their account I don't know if they're comparable U but orders of magnitude greater than what we have here across the crypto industry um here's another metric uh this one is the amount of ventor capital that has gone into crypto um since the beginning basically this chart stops at 2022 but there's another number that they care about which is the total amount of deal flow if we add this up we get a number of around 80 billion has gone into the space since 2016 so I think the interesting thing is actually when you start combining these types of data sets uh I don't have the ability to go in and look exactly how they calculate these things but I can overlay the two and when I do that I can come up with a new metric of my own and that metric is the dollars spent per developer retained and what you can see here is that for every $3.1 million that has been invested prospectively um as risk Capital we only have one retained developer in the industry still here to show for that um so I think we can we can do better and we hope to do better the point is not just that um The Venture Capital can improve um but really that perspective funding is difficult behind every one of these deals there's a team of professionals that are researching these companies and making bets and the reality is that it's often quite difficult to do this and there's an acceptance that most of these bets are not going to work out uh so perspective funding is hard retro funding is based on the principle that it is easier to identify something that has worked looking back than to predict it going forward um vital kelp designed the mechanism optimism has been rolling it out and uh it's a pretty unique way of creating markets around impact now to be clear I don't think the most uh die in the wool retr funding Maxi would say that retr funding is the only way that anything in the world should be funded but the important Point here is that right now we have a surplus of or not a surplus but a large amount of perspective funding and we need more Retros retrospective funding in the world to Signal what things we value and drive more people to build in those directions and so that's what we'll be talking about today uh at launch optimism allocated 850 million tokens from its treasury to retr funding this is a big deal and they made a promise that if you create impact on the optimism Collective you will be rewarded for that impact what that impact is how it will be rewarded TBD but that's what we're here to discuss uh now to date optimism has deployed around 60 million op or as least allocated not all that has gone out the door um which is a pretty small share of its treasury even though this is a large sum of tokens we're still very much in the early Innings and each round has been designed around a specific set of experiments hypotheses that they wanted to test round one I think was completely offchain it was a proof of concept about 50 projects no applications the selection was done I think by the op Labs team rounds two and three are where this program gained a lot of recognition they were big Mega rounds 40 million op went out in total more than 600 projects participated there were 100 badge holders um also known as Citizens uh who were voting on projects um I was a badge holder I think there a few in the room still but they were given this enormous task of reviewing all these projects and identifying which ones they thought should deserve a certain amount of funding in response to the mega rounds in in 2024 this past year um optimism uh changed the scope a little bit and instead of having these large Mega rounds they move to a system where you have smaller more tightly scoped rounds with tighter eligibility criteria and in those rounds they also tested different experiments around badge holders and guest voters and expertise and total about 18 million has gone out and there's more allocated um through the end of 2024 again we're going to be looking at the results of the program so far but putting this in the scale of other public goods funding experiments this is pretty unique it's pretty large um and my team and I at open source Observer the fact that this was a real experiment of significant scale was What attracted us to want to pay attention and work at optimism um there's a lot of economists that like writing long papers about how public goods funding is broken but here was a real chance to throw magic internet money and test whether we could actually change the incentive landscape um so I think people should pay attention to this regardless of whether they participate directly or not but first because the only thing that people like more than throwing tokens around is complaining about things on Twitter I want to address the few myths about optimism and the program of retr funding uh so the first myth is that retr funding comes with conditions the reality is the projects that receive retr funding are getting with no strings attached they can cash out they can migrate their project to salana they can delegate to governance um or they can keep building and what we see continuously is that most people actually tend to keep building on optimism uh the second myth is that retr funding is optimism's only Grand program this is also not true um it is also oop sorry it is not the only grants program it's also not the largest grants program optimism also runs prospective funding uh since 2024 they've given out more than 60 million I believe in prospective Mission based grants and most of the projects that receiving retr funding have also received some form or a large number of the products that have received retr funding have also participated in these prospective grants um I think there's a uh a narrative that optimism expects all of its teams to work for free and pray for retr funding is not the case there are various mechanisms and retr funding is just one of them and obviously the most experimental at this point myth number three optimism no longer cares about public goods you've probably heard this before if you follow the the optimism narrative on on Twitter um I think this narrative really took root after optimism changed the name of the program uh but the reality is that there was never a formal definition of what a public good is optimism and I kind of agree with this definition have always viewed it as a kind of a spectrum and the role of retr funding is to fill fill the Gap is identify where is the market not effective in rewarding impact and using tokens strategically to fill that Gap myth number four VCB projects have an unfair Advantage uh we can actually check the data on this and what we see is that uh VC backed projects are a relatively small number of the projects that have received retr funding and historically they have not performed as well as projects that do not have um any VC funding so at least to date VC projects do not seem to have any unfair Advantage even though there are some large ones that have participated in the rounds and myth number five is that optimism's badge holders are super good at deciding how much retr funding each project deserves the reality is that we are just humans um we have other jobs in many cases other ecosystems that we are that we are working on and so it is very difficult for any one badge holder to have a a God view of all of the projects and what they're doing and to be able to review each one with a level of care that that project most likely deserves um so I think that a lot of uh improving retr funding is going to be about finding the right balance of figuring out what humans and algorithms are good at so that we can make it more efficient but at the same time ensure that we're learning continuously how to improve the allocations and make sure we get closer to impact so far so good now we're going to get into the so what has this program actually delivered on these promises is it effective at rewarding imp impact um or is it just a expensive well-branded marketing campaign uh to date optimism has supported a lot of projects more than 700 um and there is a power law distribution in terms of how much funding they have received so what this graph is showing on the xais is all of the projects ranked by how much funding they have received since the start of retr funding and you can see that a average project who's been in the program has gotten about 40,000 op a top one 160 and there's a few that have received over 500,000 at the time probably worth more than a million dollars um so this has been the distribution so far it is a power law it's not a kind of flat even distribution and what you can see is that the projects that tend to receive more funding are projects that many of you may be familiar with at the very top you have protocol Guild you also have teams that have focused on developing specifically for optimism um like test and prod you have core Primitives on the super chain like um account abstraction 437 um e you also have some of the major defi protocols like vome um and social apps like Zora you also have developer libraries like ethers and weim and open Zeppelin contracts that historically have not received any funding um or of this scale and so this is also creating a recurring Revenue model for those types of projects in any case it isn't just funding a narrow set of defi protocols this is a pretty expansive and diverse set of projects that have been receiving retr funding and anecdotally um retr funding has had a fair amount of impact um one project L2 be they wrote this um beautiful letter at the end of one of the rounds thanking retr funding in the badge holders but also explaining how it actually changed how they thought about their purpose in the ecosystem um ween ethereum news uh this is also one that I can really relate with um that it was kind of a GameChanger for them receiving retr funding it allowed them to work on this and not pursue other things at open source Observer um this was definitely a GameChanger for us and gave us the confidence to leave start a new organization and focus on building for optimism um so are we a few outliers or is there some meaningful signal here at opensource Observer um the the project that I'm working on WE track cohorts of other projects and collections to see how they perform over time and one of the things that we can clearly see in the data if you remember that developer Capital report I showed at the beginning the trend line for optimism projects especially ones that have received retr funding is different on the x-axis here you have time we have a few Milestones um related to retr funding three the biggest round and you can see on the y axis the number of monthly active developers that have been participating in those in in the program both before and then coming out the other side and you know the line is up and to the right but there could be a lot of things that are explaining this so the next thing we want to look at is how would they have compared relative to some kind of a counterfactual or control group we're not in a position to basically separate out the groups and do a pure experimental test but what we can do is create a synthetic control using data about other ecosystems at open source Observer we look at not just optimism but a range of other crypto ecosystems so we can create comparable metrics for how projects in those ecosystems have have performed in the absence of receiving retr funding and so through this we can see that there's actually a significant increase relative to synthetic control for projects that received retr funding from round three excuse me we can do this for any number of metrics and I've only shown a few here but this is a a comparison of onchain activity for projects that receive funding from retr funding 4 around that Focus specifically on onchain Builders and here you can see that um is a pretty significant difference uh that has happened both um around the time of sign up but then after the results for retof funding for were were announced the challenge here is that there are other confounding factors we had the blobs excuse me um we had the the den couldn't upgrade and as a result of that transaction activity spiked across the board so we can strip that out and in this case um we don't see as clear a signal as we saw previously with the developer activity so this is a whole line of research that we need to do that we need to get better at to isolate the impact that retr funding has had on different excuse me on different cohorts of projects now if we step back and look at the overall allocation that optimism has had so far we can see that most of the funding has actually gone towards these two big boxes over here so that includes a range of governance education thank you um education analytics projects as well as technical contributions to the op stack those are these two big boxes over here and then off to the right we have some of the categories that are probably most critical to the success of the super chain which have historically received relatively less funding so that's about 14 million for the onchain builders and then about 4 million for the developer libraries so I know there's other programs that are other rounds that are in in the works right now to put more funding into developer tooling but historically more of the funding has actually gone in these other areas and so that could partially explain why there's been less impact on the uh on the onchain activity than General open source developer activity so what's next um this is just one of a number of areas that we are excited to explore together with the optimism Community obviously there is some quantitative and qualitative evidence that retr funding is having an impact but we really aren't in a position right now to make any causal statements about that impact for us at open source Observer this is a really big area of focus in 2025 we also know and you don't need to um bring up good heart's law just yet but we know that it's very easy to turn metrics into targets and to game them on the other hand we also know that to scale this program data is going to play a critical role and so the real challenge here is finding the right balance of human human judgment and data that can improve decision making um metrics should be evolutionary as we learn more information they should improve and we should be able to back test the results that we get from each of these rounds against the metrics and the decisions that we Ed to to fund those things and ideally incrementally improve each time around um this is also a big area of focus for open source observer in 2025 one of the things that we learned from recent rounds is that technical experts evaluate impact differently than non-experts now this shouldn't be a surprise to anyone um up to now the badge holder pool that optimism has assembled consists mostly it consists of a mix of experts technical experts and and non-experts and in round five there was an experiment that was done to actually test the groups of experts and non-experts prefer different projects did they vote differently the answer probably not surprisingly was was yes um there are pretty big differences in how they voted and not just on the projects they picked but also how much funding they thought top projects should receive so this chart here shows some of the most divisive projects ones that received um a pretty big difference or big spread between what experts thought they deserved and what non-experts thought they deserved um and so going forward I think that feedback is going to be used to improve the design of future rounds and particularly ones that are more technically focused ensure that the right composition of B holders are equipped to um you know uh vote on on what this product should receive um another issue that comes up are just like the incentives this uh this sends to different types of Builders uh so you have some big projects you have some small projects and when you have an allocation which is not um like a very steep power law distribution what you can what you can see is that smaller projects there might be an incentive to create smaller projects or to atomize your work into kind of a a set of very small individual contributions um as opposed to capture the collective impact of of a larger team we can see this at least in in round five where some of the projects that did the best had the smallest team um per on a per contributor basis and so this could create some kind of perverse incentive for teams to fractionalize their work as opposed to representing it as one cohesive um one cohesive team uh trend from protocol Guild a very nice essay pointing out this fact and doing some comparisons um and then the fifth one is that there is a trade-off between mathematical accuracy and just The Vibes and the spirit that comes with each of these rounds um we quants we tend to suck the life out of things and there's something very special about the spirit and The Vibes that are created around each of these retr funding rounds and so as the program becomes more accurate and predictive um there could be a trade-off in terms of the uh the energy and the attention that this gains um because there's less speculation and less hope for some for some Builders we definitely saw that round three kind of marked a high water point in terms of the social media attention at least that retr funding received um we believe that the upside to figuring out these things and working on these challenges has a potential to be huge if you just look at the size of the number of developers that are contributing to open source is a global movement and If This Were a nation it would be as large as the population of Thailand or Mexico or Germany so there is a large Global borderless movement around open source and historically we have not been very good at funding it If This Were a nation the amount of funding that it is allocating towards its public infrastructure would be on par with Fiji and swasiland pretty pretty low down the ranking so we need to do better um we at open source Observer um are very excited to work on this problem we're hoping to attract more data scientists to care about the intersection of public goods and impact and data science and if that speaks to you we'd love to hear from you you can scan the QR code to get in touch um join our Discord and for any builders that in the room optimism is having a seventh round of retr funding it'll be starting hopefully um before the end of the year you can go to retrun optimism. get all the details there and sign up thank you very much awesome Carl so you know next retr funding it's coming close so take the opportunity questions that you have for Carl let's start that way meanwhile please use the QR code to place your questions uh to e so go ahead hi Carl Thomas here hey uh thanks for the presentation I'm uh just want to say first of all I'm a huge advocate for open s so far I use it all the time my day-to-day activities always always using libraries Frameworks which are you don't even know how open source but are right um so I wanted to go back on two slides that you had uh previously me mentioned um one of them was around the development onchain activity you don't have to go back but it's fine um I was wondering would you I I know you said there were a handful of other metrics while you were speaking but you um outline these two wondering would you say these are the most important metrics for measuring an inval evaluating the impact of funded projects um no I would say that they are good trailing metrics so ideally if you have if you're doing other things well then you'll get more developers you will get more transactions you'll get more users but uh you probably don't want to fund these things directly if you were to just fund People based on the size of their team then you get very big bloated teams you wouldn't actually get the impact that you care about um on the other hand they do probably correlate in many cases with like at a high level the growth of an ecosystem um so we don't really I I'm not in a position to say there's there's one metric where you should just find this thing and I think that's that's the whole point is that we need to test these things out in the wild we need to combine them in different ways and then I think over time we'll see some kind of composite set of metrics some which are really good as these trailing indicators and some which could actually be more predictive of a future success yeah big big search space and just a followup question how much how much time would you reckon that would you would need for that for for which part for like to eval to you you mentioned you need like your Gathering data and your your to make an inform well-informed decision how how much time do you need to make that decision um we see this much more as like a community uh a challenge for the community so we can definitely come up with metrics ourself but increasingly we want to partner with with data teams um you know projects themselves people that have an intuition about what metrics they think are most useful and test those out experiment with them and you know everything all the metrics that we're building are are fully open so people can come and and test them out directly directly thank you yeah now we have some questions here one is what if what are the opportunities for data scientists to get involved yeah so um scan the QR code we have a number of contribution opportunities some are pretty basic just like helping add update um process data what we're really looking for help is people that have the ability to do kind of modeling um so you know strong python skills they're able to do things like the the causal analysis that we showed um they're able to process uh you know help help us connect new data sets and process larger and larger amounts of data um anyone who has the ability to kind of formulate research questions and then design a study around that would be super welcome um but also people that just want to uh you know create front ends and dashboards on top of it there's a lot of of cool things you can do great awesome and for for how long is optimism going to be running retro PGA fronts um I think the answer is probably forever that would be the that would be the goal um certainly the the plan is that once the the treasury the initial treasury runs out that there's a continuous source of funding that will come in from sequencer fees or from other sources of of onchain Revenue uh I think the goal eventually is that everything that is happening in optimism gets funded by retr funding whether that takes five years or 10 years remains to be seen um but the goal is that this is a system that that runs forever awesome Carl appreciate it and appre appreciate all the questions you have we just run out of time so please give a great Applause to Carl thank you everyone and optimism thank you so much [Music]