Gitcoin: What Changes When Small Donations Come from Many People?
Gitcoin’s 2019 grants experiment asked how small donors could help decide which shared tools received support. Quadratic funding offered a formula, while a limited matching pool, coordinated accounts and unequal attention tested its assumptions.
3-Minute Fast Briefing
- The ParadoxQuadratic funding gives broader support more weight by adding the square roots of individual contributions and squaring the result.
- The Turning PointIn an unscaled example, four $4 donations imply $64 total funding, including $48 from an outside pool. These are illustrative calculations, not guaranteed payouts.
- The LegacyGitcoin’s third 2019 round used a modified formula that limited subsidies generated by pairs of contributors; distinct accounts do not automatically mean independent people.
Chronological Timeline
Buterin, Hitzig and Weyl submit the first arXiv version of their public-goods funding paper.
Buterin proposes a design that bounds the subsidies generated by pairs of contributors.
Buterin publishes a review of Gitcoin’s third quadratic funding round and its modified formula.
A Developer in the Results Table
In October 2019, Vitalik Buterin reviewed a Gitcoin Grants round and found a result that made him reconsider his own priorities. Austin Griffith, known for Burner Wallet, had received $23,911 in total funding. Buterin wrote that he had paid relatively little attention to the wallet before seeing its support. A funding mechanism had made the preferences of other users harder for a prominent technical figure to overlook.[3]
The round supported 80 projects through 477 contributors and a $100,000 matching pool, according to that review. The experiment was about more than collecting donations. It asked whether people who each gave a little could help direct a larger pool toward software, education and other work available to a wider public.[3]
Why Add Square Roots?
A shared software tool can benefit people who never pay its developer. The funding design proposed by Buterin, Zoë Hitzig and E. Glen Weyl addressed that public-goods problem. Their paper, first submitted to arXiv in September 2018, described a mechanism in which the funding amount depends on the square of the sum of the square roots of contributions. The attractive efficiency result belongs to the paper’s model and assumptions, not to every real-world fundraiser.[1]
Here is a deliberately small, unscaled example. Four independent people give $4 each. Each square root is 2, their sum is 8, and its square is 64. The project’s calculated total is therefore $64: $16 in donations plus a $48 subsidy. One person giving the same $16 produces a square root of 4, which squares back to 16, leaving no subsidy in this basic example. These are our calculations of the standard formula, not historical Gitcoin payouts.[1][2]
The Extra Money Has to Come from Somewhere
The arithmetic does not mint money. The subsidy comes from a separately funded matching pool. If calculated subsidies exceed that pool, one simple budget-constrained version reduces them proportionally. A project’s actual match then depends on the available budget and the other projects as well as its own donations. The example’s $48 is an unscaled signal, not a promise that four donors can demand $48 from an empty fund.[2]
This also explains why quadratic funding is not just counting heads. Contribution sizes still enter the formula; their influence is transformed. A second person’s support adds something different from the first person giving more. The design tries to make shared benefit visible in the allocation of an outside subsidy, while retaining information about how much each person is willing to contribute.[1][2]
Who Counts as Another Person?
That distinction creates a difficult question: are the contributors independent? One person can control multiple accounts, and different people can coordinate to capture subsidies. In a June 2019 research post, Buterin proposed limiting the amount that a pair of contributors could generate across projects. The pair’s pattern of joint support would affect its subsidy coefficient. More accounts alone could not be treated as evidence that more independent preferences had appeared.[4]
Gitcoin’s third round used a modified, pairwise-bounded formula. Its results therefore should not be reconstructed as if the simple square-root example were the entire production rule. The change also had a tradeoff: correlated support might come from manipulation, but it might come from a close-knit community with sincere shared interests. The research discussion acknowledged that limiting coordination could disadvantage genuinely strong preferences too.[3][4]
A Useful Signal Is Still an Incomplete One
The October review found another limit: attention. Familiar projects could attract more contributors and receive disproportionately strong matching, even when a less visible project did comparable work. Buterin also noted weak support for Gitcoin’s own maintenance. These observations make the experiment more interesting than a claim that a formula automatically discovers the best work.[3]
Griffith’s result suggested that repeated grants might support an individual working on public tools; the review framed that as a possibility, not a guaranteed salary. Gitcoin had given small donors a way to influence funding beyond their own wallets. Whether that influence produces lasting public value still depends on real people, finite sponsor money and the work that happens after a round ends.[1][3]
Connected Stories in this Universe
Explore the chain reaction of historical breakthroughs, blunders, and legends.
The Pineapple Fund: 5,104 Bitcoins, $55 Million, and an Unrevealed Donor
Pine pledged 5,057 BTC (~$86M) on Reddit. Five months later, the fund reported 5,104 BTC ($55M) for 60 causes, leaving public donation records while remaining anonymous in the cited sources.
Read story →
Decred Politeia: After the Vote, Who Sends the Money?
A community could approve a budget while someone else still held the payment keys. Decred made that gap explicit.
Read story →Sources & References
- [1]Source 1: A Flexible Design for Funding Public GoodsVitalik Buterin, Zoë Hitzig and E. Glen WeylAccessed 2026-09-15
- [2]Source 2: Quadratic Payments: A PrimerVitalik Buterin · 2019-12-07Accessed 2026-09-15
- [3]Source 3: Review of Gitcoin Quadratic Funding Round 3Vitalik Buterin · 2019-10-24Accessed 2026-09-15
- [4]Source 4: Pairwise coordination subsidies: a new quadratic funding designVitalik Buterin / Ethereum Research · 2019-06-04Accessed 2026-09-15