How to write a Web3 grant proposal that gets funded

The 11 sections of a Web3 grant proposal

A complete Web3 grant proposal has 11 sections. Application forms rename, merge or cap them, but reviewers look for each one. The examples below marked "illustrative" are invented to show the difference in phrasing; they don't describe real projects.

1. Summary

The summary is one paragraph that says what you will ship, who uses it, on which chain, by when, for how much and under what license. It is the first thing a reviewer reads, so it has to work on its own.

• Weak (illustrative): "We are building the next-generation infrastructure layer to bring the next billion users onchain."
• Strong (illustrative): "We will ship an open-source TypeScript SDK that lets wallets batch token approvals, with docs and two wallet integrations, in 14 weeks for $38,000 over three milestones. MIT license."

2. Problem

State who has the problem and show evidence that it exists: linked issues, forum threads, support data or onchain numbers. A problem only you have described is a weak basis for a grant.

• Weak (illustrative): "Developers struggle with poor UX."
• Strong (illustrative): "Three wallet teams have open issues asking for batched approvals (links below), and approval errors were the most common topic in our wallet's support tickets over the last 90 days."

3. Solution and technical approach

Describe the architecture, the main technical choices and what already exists. The Stellar Community Fund expects a complete technical architecture before it funds a build, and the Flare Grants Program asks for methodology, architecture and a roadmap. Name the prior art and say why it falls short; MoonDAO's template asks for novelty and prior art outright.

4. Why this ecosystem

Explain why the work belongs on this chain and what the chain gets that it doesn't have today. The Solana Foundation asks why your project needs Solana rather than another chain. The Uniswap Foundation (which is spending down its remaining budget and plans to close once it is committed) wants work that could not be submitted unchanged to another protocol. Flare scores integration with its FTSO and FDC data protocols as a selection criterion.

• Weak (illustrative): "Chain X is fast and has a great community."
• Strong (illustrative): "No SDK on Chain X supports its native account abstraction, so wallets here fall back to one approval per token; ours uses it directly."

5. Team and track record

Show that you have shipped similar work, with links reviewers can open: repositories, merged pull requests, live products, past grants and their reports. Circle wants founders with a record of shipping and clear technical ownership, and Btrust's Starter Grant asks for at least one merged, non-trivial pull request. References count at some programs: OpenSats asks for two written references, Maelstrom for at least one.

6. Scope and milestones

Break the work into milestones a reviewer can check, each with a deliverable, acceptance criteria, evidence, a date and an amount. The Compound Grants Program asks for three to five milestones, mostly tied to measurable outcomes, though its main Dapps domain has been paused since April 2026 and its future is under review. Filecoin Open Grants ask for 2 to 4, although reviews there look stalled. The milestone-based grants guide shows a full table.

7. Budget

Cost each milestone from its line items. Akash's community pool asks for hours, hourly rate and other costs (though no outside team has won a pool vote there since August 2025), and proposals to Livepeer's community treasury rarely pass without a line-item budget. State the currency you priced in and the one you expect to be paid in. The Web3 grant budget guide covers rates, audits and token volatility.

8. Metrics

Promise numbers you can measure and the reviewer can check, and say where they will come from. Ink's Spark grants want claims about users, volume or TVL backed by data reviewers can verify. TheDAO Security Fund requires at least a third of each budget to sit on adoption milestones that show outside parties use the result.

• Weak (illustrative): "We expect massive adoption."
• Strong (illustrative): "Two named wallets ship the SDK by milestone 3; weekly batched approvals visible on a public dashboard; 25 GitHub dependents by month 6."

More on choosing them in grant impact metrics.

9. Risks

Name the three to five things most likely to go wrong and what you will do about each: technical unknowns, dependencies on other teams, adoption and security. Some programs make security a condition. The Tezos Foundation requires DeFi projects to pass a third-party audit of all smart contracts, and Cardano PRIME asks for a realistic audit plan for new contracts. A risks section that says "none" reads as inexperience.

10. Sustainability

Say what happens after the grant: who maintains the code, what it costs to run, and where the next money comes from. The Walrus Foundation looks for a path to sustainability such as other funding, and the Canton Development Fund asks for a plan for who maintains the output. See funding after the grant ends.

11. Open source and license

State the license and when the code becomes public. Requirements differ: Filecoin Open Grants require dual MIT and Apache-2 licensing, the Monero CCS requires a permissive license at every stage, and the Stellar Community Fund asks for a plan to open-source any smart contracts. Rocket Pool's committee asks you to say whether the results will be open source. See open source licenses for grants.

How should a grant proposal be formatted?

Format a grant proposal so a reviewer can find any answer in seconds. Popular programs get hundreds of applications per round (the Stacks Endowment's first cycle drew 325), and a clear layout is the cheapest way to look competent.

• Match the headings to the form. If the program asks for "Impact", call the section "Impact", even if you would have called it "Metrics".
• Put milestones and the budget in tables. One row per milestone, with deliverable, date and amount; the budget table should add up to the same total.
• Link, don't attach. Repositories, demos, dashboards and contract addresses as links a reviewer can open from the form.
• Use absolute dates and one currency. "Milestone 2: March 15, 2027, $12,000 (paid in the program's token)" beats "Q1, 12K".
• Keep the numbers consistent. The amount in the summary, the budget table and the form field must match; a mismatch is the easiest error for a reviewer to spot.
• Write plainly. Short sentences survive translation tools and skim reading; jargon and adjectives don't add evidence.

How do you adapt a proposal for foundations, DAOs and RFPs?

Keep one full version of the proposal and adapt it to three formats: a foundation form you fill in privately, a DAO forum post that the public reads and votes on, and an RFP response that answers someone else's specification.

Foundation application forms

Foundation forms cap your length and fix the order, so write the full version first, then cut to the fields. Formats vary: OpenSats wants about one page, two reference letters and a video of exactly two minutes; the Stellar Development Foundation's academic grants take a PDF of at most three pages; the Tezos Foundation wants a single PDF with team, technical plan, roadmap and full budget. Ink's Spark grants have no pitch deck, while Nosana expects a pitch deck, demo and roadmap.

DAO governance forum posts

A DAO proposal is read by token holders and delegates who may not be technical, and it is debated in public before anyone votes. Open with a short summary, put the ask and the deliverables near the top, disclose conflicts of interest, and answer every question in the thread. CoW DAO wants its grant template posted on its forum, then a Snapshot vote in which 3 of 4 committee members must approve. Zcash Community Grants give the community at least a week to comment. RootstockCollective asks for one to three weeks of forum feedback and a live public AMA before a 7-day on-chain vote, and Polkadot's treasury guidance is to leave a draft open for feedback for at least 7 days. The DAO funding proposal guide covers this format.

RFP responses

An RFP response answers a request the program wrote, so mirror its structure and show how you meet each requirement. The Ethereum Foundation's Ecosystem Support Program separates Wishlist items, where you propose the approach, from RFPs with set deliverables and an application window; as of October 2026 none of either is open. TheDAO Security Fund lets you propose an RFP, which teams compete for once it is funded, or a grant for a named team with a head start. Where an RFP has an owner, talk to them first: Compound asks applicants to contact the domain allocator early to agree on milestones. More in Web3 RFP grants.

Should you use AI to write a grant proposal?

Use AI to tighten your own draft, not to write it. The Compound Grants Program says it rejects low-quality or LLM-generated proposals, and generic text fails the fit test at every program because it could describe any project on any chain. Never let a tool invent metrics, users or partners. The guide to AI grant proposal writing covers safe uses.

Before you submit: a five-minute review

Read the proposal once as the reviewer would, with the program's page open beside it:

1. Does the first paragraph name the deliverable, the users, the date and the amount?
2. Does every claim about users, traction or past work have a link?
3. Does each milestone have acceptance criteria, a date and an amount?
4. Do the milestone amounts add up to the total, and is the total inside the program's range?
5. Did you use the program's template and fill in every field?

Then run the full grant application checklist, rehearse the questions grant reviewers ask, and compare your draft with the common reasons why grant applications get rejected. The Web3 grant proposal template has all 11 sections ready to copy, and how to apply for Web3 grants covers the steps around the proposal. Find open programs in the Web3 grant directory.

FAQ

How long should a Web3 grant proposal be?

A Web3 grant proposal should be as long as the program's format allows and no longer. OpenSats asks for about one page, the Stellar Development Foundation's academic grants allow at most three pages, and Kusama's Vision art bounty accepts a PDF of up to 15 pages. When there is no limit, two to four pages plus a milestone table and a budget table is a reasonable target.

Should a Web3 grant proposal include a pitch deck?

A Web3 grant proposal needs a pitch deck only when the program asks for one. Nosana expects a pitch deck, demo and roadmap, Flare asks for a pitch deck with its form, and Kaia's IOK Fund requests one with its funding form. Ink's Spark grants have no pitch deck at all. Grant reviewers look for scope, milestones and costs, which a deck rarely shows well.

Do Web3 grant proposals have to be in English?

Some Web3 grant programs require proposals in English, and a few accept other languages. The Ethereum Foundation's PhD Fellowship required proposals in English (its 2026 call has closed and no 2027 call is announced), and Kusama's Vision art bounty requires English. Nervos's CKB Community Fund DAO accepts proposals in English or Chinese. If English is not your first language, plain sentences, clear tables and working links matter more to reviewers than polished prose.

Can you reuse a Web3 grant proposal for another program?

You can reuse the research and structure of a Web3 grant proposal, but rewrite the ecosystem fit, milestones and metrics for each program. The Uniswap Foundation, which is spending down its remaining budget, wants work that could not be submitted unchanged to another protocol. Disclose other applications for the same work: Tezos Commons asks about them directly, and VeChain's low-activity VeBetter grants allow other funding if it is disclosed.

Should a Web3 grant proposal ask for the program's maximum amount?

A Web3 grant proposal should ask for what the costed work needs, not the program's maximum. Rocket Pool's committee, short of funds, approved nothing above $7,500 between May and October 2026, though requests went up to $100,000. Lido's grants need more approvals as amounts grow: one committee member can approve up to $1,000, two members up to $10,000, and the Council up to $100,000.