Summary
The Outcomes Based Resource Allocation (OBRA) framework was ratified by SafeDAO in SEP 8, and its initial Strategies and Budgets through SEP 10. The framework, in its first iteration, started accepting proposals from January 2024. It ran until December 2024 when it was modified by the DAO introducing the Committee and Council structure through SEP 49, effectively pausing OBRA until the committees and council structure is in place.
This post looks at the ~12 months of OBRA (referred hereon as OBRA Wave 1) as the DAO prepares for another season of OBRA (referred hereon as OBRA Wave 2) to surface insights and catalyze a discussion around potential improvements.
OBRA Wave 1: Design
OBRA streamlines funding requests to SafeDAO through strategies. Strategies provide a high-level approach and key metrics to achieve the vision and goals that are stipulated in the SafeDAO constitution. Initiatives (projects proposed by potential grantees) must be driving forward existing DAO strategies. Generally, an initiative cannot receive funding unless it is tied to a pre-existing strategy.
OBRA Wave 1 debuted with 6 strategies, with the following budgets:
-
Strategy 1: Research and implement Safe token utility | Initial budget = 200, 000 USDC
-
Strategy 2: Foster module ecosystem | Initial budget = 300,000 USDC
-
Strategy 3: Increase awareness of Safe Ecosystem | Initial budget = 250,000 USDC
-
Strategy 4: Research decentralization of Safe tech stack | Initial budget = 200,000 USDC
-
Strategy 5: Increase governance participation | Initial budget = 150,000 USDC
-
Strategy 6: Wildcard strategy | Initial budget = 100,000 USDC
The grant program followed a fairly straightforward structure, where any person or organization could make an application by drafting a public forum post. The proposal then follows the normal governance processes at SafeDAO: signalling by delegates and guardians => Snapshot voting => Onboarding. The KYC, grant agreement, and milestone review are handled by the Safe Ecosystem Foundation (SEF).
Key Facts
- OBRA Wave 1 has allocated a total of 1,050,000 USDC to 25 initiatives. A breakdown of the funding flow by category is given below. Note that strategy budgets were amended through SEP 28, hence the difference from the initial budgets specified in the previous section:
As visible from the table above, strategy 2: foster module ecosystem, and strategy 5: increase governance participation allocated the most amount of funds at 355,000 USDC and 294,000 USDC respectively. Strategy 4: research decentralization of Safe tech stack and strategy 1: research and implement Safe token utility allocated the least amount of funds at 50,000 USDC and 85,000 USDC respectively. The remaining two strategies fall between these levels.
- The average grant size for OBRA Wave 1 was 42,000 USDC. Average grant sizes differ significantly between strategies:
The average grant size (total amount allocated / number of initiatives) is the lowest for strategy 3 at 27,300 USDC and goes all the way up to 71,000 USDC for strategy 2. Note that these values are purely coincidental, and no programming was involved in ‘controlling’ grant sizes.
- 5 out of the 15 grantees were awarded multiple initiatives:
Note that multiple initiatives were joint-efforts including more than one organization. The table above does not make that distinction, and hence Total Funding should not be interpreted as total funding allocated to the organization, but as total funding for initiatives the organization is involved in.
From the above table, 673,500 USDC or 64% of the total funding was allocated to initiatives proposed by 4 organizations (Gnosis Guild, Kolectivo, Areta & Curia).
- OBRA Has Dominated Governance: Since SEP 10, 42 proposals have proceeded to vote on SafeDAO’s Snapshot. Of these, 33 are OBRA proposals. To put it in perspective, around 4 in every 5 proposals in the last 12 months has been OBRA.
It is very clear that a significant percentage of governance activity at SafeDAO in the past year was grant allocation through OBRA.
All of this data is publicly available. It can also be accessed through this Google Sheet.
Discussion
1. Impact Measurement
Poor impact measurement is an ecosystem-wide problem that reduces the effectiveness of capital allocated. Without a robust framework for impact measurement, the DAO is ‘forced’ to almost always price grants based on their ‘expected impact’, even though a grantee is only bound by their initiative’s ‘outcome’. In such a scenario, the DAO will almost always overpay a grantee.
For example, where the grant agreement is effectively to deliver content, or create a module, but priced based on “the number developers it can onboard to Safe”, or "the amount of TVP it can route through Safe”, it is important to have a flexible capital allocation framework that can factor in the impact generated. This will attract higher quality projects to the Safe ecosystem, and ensure they are rewarded more if they create more impact. Introducing a KPI-triggered SAFE bonus, or separating the grant into multiple phases of outcome and impact might be interesting solutions to consider here.
2. Strategies
As strict KPIs were not attached or measured, it is not easy to discuss strategy performance.
Strategy 1 (research and implement Safe token utility), and strategy 4 (research decentralization of Safe tech stack) received the least amount of interest from grantees. This can mean multiple things, including a lack of participants in the Safe ecosystem capable of tackling these strategies, to the infeasibility of delivering value through these strategies, to the SEF needing to internalize the responsibilities defined by these strategies, or even a miscommunication on what is expected. A continuous evaluation and adjustment of strategies by an expert body (Strategy Steering Committee), as ratified by SEP 49 can possibly address this issue, and ensure strategies stay relevant to the grants program.
Strategy 2 (foster module ecosystem), while allocating the most amount of capital, has not produced a clear ‘winner’ for SafeDAO when looked at from the perspective of transaction volume or amount of usage. The reason this point is important to consider is because strategy 2 is the primary vehicle for SafeDAO to grow key metrics like TVP, and TVL and should technically be designed to catalyze more onchain innovation on Safe.
Strategy 3 (increase awareness of Safe ecosystem) had some interesting initiatives, including Superchain Eco that led to the deployment of over 40,000 Safe’s on Optimism. It might be worthwhile to further refine this strategy to ensure it can support more initiatives around marketing. Note that organic marketing in general can have a much lower cost than development. For example, high-quality events, hackathons, content, research articles, and fun experiments that increase the awareness of Safe can be funded by grants lower than 1000, 5000 or 10000 USDC. Strategy 3 is hence very different from the other five, especially in its ability to create and fund a wide variety of experiments that may not cost a lot. SafeDAO could try to refine this strategy to a vehicle that can support a broad spectrum of marketing initiatives of all sizes.
Strategy 5 (increase governance participation) was one of the most active strategies, funding 9 initiatives and disbursing the second most amount of capital. However, the governance participation in SafeDAO has not increased. In fact, it has gone down from early 2024 levels. The maximum amount of interest that can be fostered is limited by the surface area of what SafeDAO can technically govern, but we observe a gradual decline in governance activity as mindshare moved from redistributing the airdrop to token transferability and resource allocation to other topics. As mentioned earlier, OBRA has more or less dominated governance, and this might also be a potential reason for the decrease in governance interest. However, governance activity is also low on other requests that should be of wider interest (e.g. governance changes), which is something that we need to tackle.
Note that most proposals funded through strategy 5 were aimed at streamlining governance operations and increasing transparency. While these initiatives are important, SafeDAO has not been able to focus on the primary intention of this strategy, which is to improve the amount, and quality of governance participation. As OBRA serves as a support vehicle for both growing the Safe ecosystem, as well as funding its operational legos, we may need to introduce clearer distinctions between initiatives, so that the ‘primary intention of a strategy’ is not sidelined. More strictly defining KPIs, introducing an RFP process, etc. might be interesting solutions to explore on this front.
Strategy 6 (wildcard), while originally intended as a catch-all field to fund initiatives outside the scope of strategies 1 to 5. However, towards the end, it also became a strategy to fund experiments when the corresponding strategy budget was not adequate (spill-over for other categories). It is important to discuss whether this was intended and desirable in the long run.
3. Participants and Grant Sizes
While not explicitly defined, OBRA is perhaps the most effective vehicle SafeDAO has to attract new teams and builders to the Safe ecosystem. However, reviewing the funding flow, it is clear that the DAO has not had a high influx of ‘new grantees’. In fact, a large amount of funding has been allocated between a handful of projects. While this is not inherently bad, and can be considered as a sign of loyalty and proportional to the value they create, it is important the DAO manage this dynamic well to find a balance between new ideas, and continued support of existing participants. It is also important to ask ourselves if the OBRA Wave 1 being operationally very governance heavy made it more difficult for new grantees to even make it to the voting stage .
OBRA has in general not been great at funding small grants. At the moment, a potential grantee has no incentive to scale down their request, plus the ‘lift’ for getting an initiative funded is huge. By introducing a different set of incentives, for example a lighter or a shorter review process for grants under a threshold, SafeDAO might be able to catalyze a lot more experimentation and pilot programs. This might also be a better entry point for new teams and grantees to the Safe ecosystem.
Conclusion
As with any system design problem, a lot can be done to improve OBRA. Many of the issues identified above are easy fixes, and many are deeper issues that will take longer to solve. Nevertheless, the infusion of expertise and lightness into the grants stack at SafeDAO through the committees and councils Structure will continuously optimize OBRA. With enough effort, honest conversations, and iteration, we can grow OBRA into one of the top grant programs in the ecosystem.
As delegates, DAO participants, and potential grantees, your feedback is the most important input to this discussion. Please take some time to share it!