The Williams Project

The organization over time

Each line starts at 100 in 2018, so what you read is the shape rather than the size: 150 means half as much again as 2018, 50 means half. The number beside each label in the key is where it ended. The shaded band is where the middle 50% of its peers' revenue would sit, given their growth.

How it's funded, over time

Each bar is one year's revenue split into where it came from, and every bar is the same height — these are shares, not amounts, so a year that raised twice as much looks the same size. Hover a bar for the split.

Government-grant reliance: 2021 11% · 2022 31% · 2023 9% · 2024 6% · 2025 24%. Grants only. Government contracts and fees sit inside program revenue.

Surplus & reserves

Operating surplus or deficit each year, and months of liquidity in hand. 4 of the last 8 reported years ran a deficit.

Who funds it, year by year

2017 → 2023: the base broadened from 1 funder to 8 funders, grant income rose $36k → $84k.

8 of 16 of your funders are donor-advised or pass-through sponsors (tagged DAF) — 72% of grant dollars received. That money is really individual donors directing a sponsor; the institutions to cultivate are the non-DAF funders.

From the IRS filings of The Williams Project’s funders (the co-funder graph). Association, not causation. Trends are read to FY2023, the last fiscal year that has finished arriving. FY2024–FY2026 are still being filed and are shown unshaded; they do not move any figure on this page.

How concentrated its funding is

The Williams Project has a broad base — no single funder exceeds 27% of grant income, and it takes 3 funders to reach half.

95% of the income these shares are computed over arrives through pass-through sponsors or from payers whose filings do not say what kind of payment it is. Concentration is still measured over all of it, because the money is real; what it does not support is a claim about how many institutions have chosen to fund The Williams Project.

Largest funder’s share by year: 2017 100% · 2018 100% · 2019 95% · 2020 56% · 2021 43% · 2022 28% · 2023 28% — diversifying over time.

“Effective funders” = inverse Herfindahl index (1 / Σ shareᵢ²) — the number of equal-sized funders that would give the same concentration. Trends are read to FY2023, the last fiscal year that has finished arriving. FY2024–FY2026 are still being filed and are shown unshaded; they do not move any figure on this page.

How long its funders stay

43% of The Williams Project's funders are still giving 3 years after their first grant; 75% give in more than one year at all.

Share of funders still giving k years after their first recorded grant, pooled across every acquisition cohort. A funder counts as retained in a year only if it made a grant that year. Measured to FY2023, the last fiscal year that has finished arriving — a funder cannot be counted as lapsed in a year most filers have not reached.

Where its funders are

72% of The Williams Project's grant income arrives through donor-advised or pass-through sponsors, whose addresses record where the money is held rather than where the donor is. Of the $119k that is directly attributable, 53% of it comes from funders outside Washington, across 4 states.

Funder states come from each funder’s own filing. $306k arriving through sponsors registered in 6 states is excluded from the map and the split above: a sponsor holds money on a donor’s behalf, so its registered address would place those dollars somewhere no donor need ever have been. Trends are read to FY2023, the last fiscal year that has finished arriving. FY2024–FY2026 are still being filed and are shown unshaded; they do not move any figure on this page.