Skip to content
Regen Web3 Toolkit

Writing Grant Proposals

You have a node idea and a blank Gitcoin application form in front of you. The form looks easy to fill out, but the words you put in it are what decide whether your node walks away with $200 or $20,000, and whether you build something a real community actually wants.

This guide is about writing those words well. It covers the two kinds of grants you’ll meet, what a working proposal actually contains, why your proposal is really a community campaign, and how to take the money without letting it quietly take over your life.

Two kinds of grants, two kinds of proposals

Section titled “Two kinds of grants, two kinds of proposals”

Web3 grants split into two types, and the proposal you write is very different for each.

Prospective grants fund a promise of future work. You describe what you’ll build, and a committee or a crowd decides whether to bet on you. Most grants you’ll apply for as a new node, like Gitcoin rounds, DAO grants, and foundation grants, are prospective.

Retroactive grants reward impact you’ve already delivered. The clearest example is Optimism’s Retro Funding, which pays projects for public goods they shipped in the past. The thesis, from the 2021 post that introduced the idea (an Optimism blog post with Vitalik Buterin as guest author of the key section), is simple: “it’s easier to agree on what was useful than what will be useful.”

This distinction changes your writing completely. For a prospective grant, you make a credible case for a future. For a retro grant, you never write aspirational language at all; you document what you already did, with metrics. Know which one you’re applying to before you write a single sentence.

One more term, because most early-node funding flows through it. Quadratic Funding (QF) is a matching system where the number of people who back you matters more than how much each one gives. The match is calculated by taking the square root of every contribution, summing those roots, then squaring the total, so a hundred people giving $5 pulls far more matching money than one person giving $500. Broad support beats a few whales.

That math should change how you think about your proposal. In QF, the job isn’t only to impress a panel. It’s to turn out a lot of real people.

What a working proposal contains

Section titled “What a working proposal contains”

Gitcoin publishes its own guidance on this, and the patterns are the same whether you’re writing for a crowd or a committee.

Lead with a TL;DR. This is the single most useful rule in the whole guide. Gitcoin’s advice is blunt: many donors only read the beginning of a proposal, so open with a brief overview of what you’re trying to do and, crucially, how you intend to spend the money. Your first line is the hook. Make it state, plainly, what the money buys.

Metrics over promises. Be specific about impact and give both qualitative and quantitative numbers, like users, attendees, transactions, trees planted, people onboarded. For retro grants this is mandatory; for prospective ones it’s what separates you from every vague “we’ll build community” application in the pile.

Show your team. Gitcoin tells applicants to list key project members and their qualifications, because the people involved are just as critical as the innovation itself. Name the people doing the work and why they’re the ones to do it. Credibility is a person, not a paragraph of adjectives.

Iterate in public. Here’s the antidote to the fear that you’ll build something ineffective. Gitcoin tells applicants to show they’re learning from their mistakes and listening to feedback, because people will want to support a team that’s iterating, and you don’t have to have everything figured out ahead of time. A node that’s transparent about what didn’t work beats a polished one that hides everything. You don’t need to be impressive yet. You need to be visibly learning.

For local nodes: your proposal is a campaign

Section titled “For local nodes: your proposal is a campaign”

Here’s the part the form doesn’t tell you. For a local node in a QF round, the proposal and the fundraising campaign are the same thing.

Because matching rewards the breadth of your support, the grant page becomes a rallying point. You don’t raise money by polishing prose and waiting. You raise it by running events, in person and online, that get your local community to go donate, even $5 each. The proposal is the story; the events are how people find it.

The clearest example is ReFi DAO’s Local Node round on Gitcoin (GG18, 2023). Aspiring nodes joined a three-month cohort, drafted their grant pages, then fundraised during the round by hosting gatherings to rally their local communities. With a $30,000 match pool (funded by Celo, Celo Europe DAO, and KNKT DAO), the round saw 18 new local nodes emerge worldwide and collected over 1,550 contributions totalling $6,700+ in additional donations.

Sit with those numbers. The donations averaged a few dollars each. That’s the point: 1,550 small gifts, not 15 big ones. And real nodes came out of it. ReFi Medellín went on to launch a community lending platform that pools local resources for small-scale projects; ReFi Costa Rica started a Spanish-language podcast and incubator. None of them started big. They started by getting a roomful of actual people to care.

This is also your answer to the fear of building something ineffective. You don’t validate a node with a grant committee. You validate it with twenty neighbours who showed up and chipped in. If you can’t get twenty people to give $5, the grant was never the missing piece, and that’s useful to learn cheaply.

Don’t let grants become a trap

Section titled “Don’t let grants become a trap”

Now the harder part, because this is where nodes burn out.

A grant feels like a finish line. It’s not. Single-source grant funding is one of the most reliable causes of burnout and turnover in mission-driven work: the money is unstable, the round ends, and you’re back at zero, writing the same application while trying to keep events running. Funders know this too. They don’t want to fund you forever, and most expect you to rely on a decreasing share of grant money over time.

Grant-writing guidance puts the goal plainly. The sustainability section of a strong proposal should “describe a plan to rely on a decreasing portion of grant support over time by boosting support from other sources.” That’s not a line for the application; it’s a survival strategy for you.

So diversify from the start. Greenpill Network’s local chapters, like Greenpill Denver and Greenpill NYC, sustain themselves through recurring local and network-level QF rounds and other mechanisms rather than betting everything on one big grant. ReFi DAO ran a structured round for many nodes, not one prize for one node. The pattern in both: lots of smaller, repeatable inflows beat one large pool you become dependent on.

This is also your defence against central overreach. A node that gets all its money from one central pool answers to that pool. A node funded by its own community, several grants, and a bit of its own revenue answers to itself. Owning your funding base is the same thing as owning your independence. The broad-support math of QF isn’t an accident of the mechanism; it’s pointing you at the healthiest way to be funded.

Try This

Section titled “Try This”

Start here: Write a three-sentence TL;DR for your node, right now. Sentence one: what your node does. Sentence two: exactly how a grant gets spent. Sentence three: the impact, with at least one number you can actually count. If you can’t write sentence two concretely, that’s the work to do first.

Go deeper: Draft a real Gitcoin grant page for your node, and before any round opens, line up 20 community members who’ll donate. Tell them now: wallet setup and onchain donating take time, and people who fund you on the last day usually don’t fund you at all. Twenty committed small donors is a stronger signal than a perfect application.

Stretch: Design a 12-month funding mix where grants are less than half your total support. Map the other half: community contributions, a recurring QF round, partner funders, modest node revenue. Write down what you’d do the month after your biggest grant ends. If that month looks fine, you’ve built a node, not a grant dependency.

References

Section titled “References”