Gitcoin
Direct Grants

Direct Grants

A funding allocation mechanism where expert reviewers allocate capital based on judgment, strategy, and expected impact.

Direct grants are a capital allocation mechanism in which a designated individual, committee, or domain expert evaluates proposals and awards funding based on strategic priorities, technical merit, and expected impact. Direct grants are the most widely deployed funding mechanism in the Ethereum ecosystem, forming the operational backbone of nearly every protocol foundation and DAO treasury program.

Unlike community-driven mechanisms such as quadratic funding or conviction voting, direct grants delegate allocation authority to reviewers with relevant domain expertise, trading democratic signal for depth and speed of evaluation. Many programs structure disbursement around milestones, releasing capital incrementally as grantees deliver on agreed deliverables.

How It Works

Many funding needs — infrastructure research, security audits, and protocol development — require deep technical evaluation that broad community voting cannot reliably provide. Community-driven mechanisms excel at surfacing popular projects but can underweight specialized, unglamorous, or long-horizon work.

Direct grants address this by delegating allocation authority to reviewers with relevant domain expertise, who evaluate proposals against defined criteria and make funding decisions. This approach trades the democratic signal of crowdfunding for the depth and speed of expert judgment. Direct grants are frequently composed with milestone-based disbursement as an execution layer.

  1. Strategic scoping: A foundation, DAO, or treasury defines funding priorities, often organized by domain (e.g., developer tooling, security, education, research). Some programs publish explicit wishlists or Requests for Proposals (RFPs) to signal areas of need.
  2. Proposal submission: Applicants submit proposals through an application process — typically a form or onchain submission — describing the project, team, methodology, timeline, budget, and expected impact. Some programs accept open applications; others operate by invitation or proactive scouting.
  3. Expert review: Designated reviewers — foundation staff, elected committee members, or domain allocators — evaluate proposals against published criteria. Review processes range from single-reviewer decisions for small grants to multi-stage committee deliberation for larger awards.
  4. Funding decision: Reviewers approve, reject, or request revisions to proposals. Approved grants specify a total award amount, disbursement schedule, and deliverable milestones. Many programs require KYC verification and legal agreements before disbursement begins.
  5. Milestone-based disbursement: Capital is released incrementally as grantees complete agreed milestones. Grant evaluators conduct regular check-ins and milestone reviews to verify progress. This staged approach limits downside risk and maintains accountability throughout the grant lifecycle.
  6. Reporting and evaluation: Grantees publish results — code repositories, research papers, public reports — upon completion. Programs may track outcomes to inform future funding decisions and build institutional knowledge about what works.

Advantages

  • Expert-driven capital allocation: Funding decisions are informed by domain knowledge and strategic judgment, enabling support for work that requires technical evaluation beyond broad community voting.
  • Prioritization of low-visibility but high-leverage work: Core infrastructure, security research, and long-horizon protocol development can be funded even when they lack public visibility or a natural donor base.
  • Depth over breadth in evaluation: Reviewers can assess feasibility, architecture, and execution risk in ways that lightweight or popularity-based mechanisms cannot.
  • Accountability through staged execution: Milestone-based disbursement ties capital release to verifiable deliverables, reducing downside risk throughout the grant lifecycle.
  • Flexible support across the project lifecycle: Direct grants can fund work from early-stage research to scaling proven tools, with grant structures tailored to context.
  • Scalable delegation via domain allocators: Review authority can be distributed across multiple experts managing parallel funding tracks, increasing throughput while preserving subject-matter depth.
  • Organizational neutrality: The mechanism functions across centralized foundations, hybrid governance bodies, and DAO treasuries.

Limitations

  • Committee capture and bias: Concentrated decision-making authority creates risks of groupthink, favoritism, and blind spots. Reviewers may systematically overlook work outside their expertise or networks.
  • Accountability gaps: Without transparent evaluation criteria and public reporting, it can be difficult for communities to assess whether grant capital was deployed effectively.
  • Scalability constraints: Expert review is labor-intensive. As application volume grows, review quality can degrade or processing times extend, frustrating applicants and slowing ecosystem development.
  • Applicant burden: Detailed proposals, milestone reporting, and KYC requirements create overhead that disproportionately affects small teams, solo contributors, and builders outside institutional networks.
  • Principal-agent misalignment: Reviewers may optimize for their own priorities or institutional incentives rather than ecosystem-wide impact, particularly when feedback loops between funding and outcomes are weak.
  • Limited community signal: Unlike quadratic funding or conviction voting, direct grants do not generate public information about which projects a broader community values, potentially missing grassroots priorities.

These constraints make reviewer selection, transparency, domain diversity, and feedback mechanisms the central design challenges for direct grant programs.

Best Used When

Direct grants work best when:

  • Funding decisions require technical or domain expertise
  • The work is strategically important but lacks broad visibility or donor support
  • A trusted entity can define priorities and evaluation criteria
  • Milestone-based accountability is needed to manage risk
  • Speed of deployment matters
  • The goal is to complement community-driven mechanisms by covering funding gaps

Examples and Use Cases

Ethereum Foundation Ecosystem Support Program (ESP) is one of the longest-running direct grant programs in the ecosystem, launched in 2018 to support open-source public goods across developer tooling, cryptographic research, infrastructure, and community development. In late 2025, ESP shifted from an open application model to a proactive, wishlist-driven approach organized around Requests for Proposals, reflecting the scalability limits of committee review and a move toward tighter strategic alignment.

Arbitrum DAO Grants operate through a Dedicated Domain Allocator (DDA) model, in which community-elected domain experts manage grants within specific focus areas using delegated budgets and independent review authority. This structure decentralizes decision-making while preserving expert judgment, demonstrating how direct grants can scale through pluralistic review rather than centralized committees.

Polygon Community Grants use a domain allocator structure spanning DeFi, AI agents, consumer applications, infrastructure, research, and community development. Allocators review proposals against published rubrics, with milestone-based payouts tied to deliverables, illustrating direct grants operating at protocol-ecosystem scale with structured governance oversight.

Base Builder Grants take a scouting-first approach: rather than accepting formal applications, the Base team proactively identifies and evaluates live products based on adoption, novelty, and onchain impact. Small, discretionary grants awarded after direct product review demonstrate that direct grants can operate as a lightweight, high-signal allocation mechanism without a traditional application pipeline.

Further Reading

Tags

expertmilestonegrants
Edit on GitHub

Updated: 2/13/2026