Journal

How to write a milestone a grant committee can approve

By the Thaw Ventures team · 2026

Most grant applications are decided in the milestone section. The thesis says what you believe; the milestone says what the money buys. If a reviewer cannot see that in a few lines, the rest of the application rarely makes up for it.

Four things a reviewer checks

• One verifiable deliverable. A single thing that either exists or does not: a contract on testnet, a library release, a hosted demo. Not an activity such as “work on” or “explore”.
• A date. A calendar date, or a number of weeks from funding.
• Budget lines. What this milestone costs and nothing else, in a few lines that add up to the amount asked.
• An acceptance test. A link, a command or a script that anyone can run to get a yes or a no without asking you anything.

The acceptance test is the part most people skip, and it does the most work. If you cannot write it, the deliverable is not concrete yet.

A before and after

Imagine a team building a small app that rebalances a wallet between three tokens, following rules the owner sets.

Before: “Milestone 1: research and build the core architecture and integrate with DeFi protocols. Budget: $40K.”

After: “Milestone 1, six weeks from funding: the app runs on a public testnet against one exchange, rebalancing a test wallet between three tokens according to a rule file the owner edits. Spending limits are enforced by the wallet contract. Budget: $18K, split into engineering, hosting and an external review of the contract. Acceptance: a public link to the deployed contract and its transaction log, plus a test script that tries to exceed the spending limit and fails.”

The second version is smaller, names one integration, and answers the question a reviewer will ask anyway: what stops the app from draining the wallet?

Common mistakes

• Research with no output. Research is fundable when it produces something: a report that answers stated questions, a benchmark, a prototype.
• Bundling. Three components in one milestone means one late piece makes everything late. Split them.
• Outcomes you do not control, such as user numbers or exchange listings. Tie the milestone to what you ship.
• A round-number budget with no breakdown behind it.

Size the milestone first

Choose the milestone, then let the amount follow from its budget, not from a program’s maximum. Asking for the top of a range for a first step tells a reviewer the work has not been scoped.