Plain-language summary: Funders sort nonprofit spending into three buckets — program, administration, fundraising. This page says exactly how we sort ours, why our two paid staff count as program rather than overhead, why our fundraising line is zero, and what our real numbers are. In 2025, 88% of what we spent went to program. It also says what that number hides.
Table of Contents
- Purpose
- Three Categories Built for Someone Else
- Program Expenses
- Administration Expenses
- Fundraising Expenses
- Under Review: the Fourth Category
- How We Classify
- What Our Numbers Actually Are
- What the Ratio Cannot Tell You
- On Overhead
- When We Get It Wrong
- What Changes When We Become a Public Charity
- Who Is Responsible
- Review of This Policy
Purpose
Nonprofits are judged by how they spend money. The BBB Standards for Charity Accountability and most institutional funders ask for spending sorted into three categories: program, administration, and fundraising.
This page sets out the rule we apply to every line in our books, so that anyone reading Where the Money Went can check our arithmetic rather than take our word for it. A ratio without a stated method is a marketing number. We published one of those for years, and it was wrong.
Worth saying plainly: this three-way split is not something our tax filing requires. As a private foundation we file Form 990-PF, which does not ask for it. We report it because funders and evaluators do, and because we would rather answer the question in public than be sorted by someone else’s guess.
Three Categories Built for Someone Else
Program, administration, and fundraising was designed for organizations that hold galas, buy advertising, and employ development staff. We do none of those things.
Forced into those three boxes without thought, our spending produces a number that is wrong about us in the direction of looking worse. That is not a hypothetical. Our previously published figure said 65% of expenses went to program. The real figure was 88%. We had been marking ourselves down by more than twenty points, and reporting a fundraising function we do not have, because the framework could not see what our staff actually do.
So we use the categories — they are useful, and refusing them would just make us harder to trust — but we say out loud where the fit is bad.
Program Expenses
The direct costs of doing the mission. For us that is:
- Mutual aid grants and direct support to Disabled and Neurodivergent people
- Creator grants and community compensation
- Operations staff — our two paid people
- Software, apps, and hosting
- Programs and events
Two of those need defending, because they are the two most organizations would file as overhead, and together they are the great majority of our spending.
Our two paid staff are the program. They run the grant programs and deliver the non-monetary mutual aid — finding services in people’s own localities, arranging healthcare and social services appointments, helping people navigate systems designed to exhaust them, and checking in afterward. Much of that work is done alongside people in genuinely perilous situations, many of them without housing. Counting it as administration would describe the core of this organization as overhead.
Software is program tooling. It hosts and maintains our public knowledge gardens, and it provides software and internet access to members of our community. Internet access for a Disabled person who could not otherwise afford it is mutual aid, not office overhead.
Administration Expenses
The costs of being a lawful, insured, auditable organization. Reported as Management & General in our books:
- Accounting and bookkeeping
- Taxes and required filings
- Insurance
- Licenses and state registration
- Legal and governance costs
- Bank and investment fees
- Shipping, postage, office supplies, and printing
- Memberships and subscriptions
This is a small list and a small number, and neither is an accident. We have no office, no vehicles, and no development department. What is here is what it costs to stay legal, insured, and checkable — including the accountant who found the errors we have had to publish corrections for.
Fundraising Expenses
We do not fundraise, and our fundraising line is zero.
No events, no advertising, no mailings, no development staff, no consultants working on commission. Our 2026 fundraising figure is $0, and that is the accurate description of our practice rather than an accounting choice.
If that ever changes — if we run a campaign, hire someone to raise money, or pay a platform specifically to solicit — those costs are booked here and this paragraph gets rewritten. Donation platform fees on gifts people choose to give us are not solicitation, and we say so rather than quietly parking them somewhere flattering.
This has a mundane consequence worth naming: an organization that spends nothing on fundraising raises very little. See below.
Under Review: the Fourth Category
We use a fourth category that the standard framework does not have: Under review. It holds spending we have not finished tracing.
In 2025 it holds $3,570 posted to a fundraising fees account. Our current understanding is that it paid for merchandise, printed materials, and event signage — which would make almost none of it fundraising. We have left it where it was booked rather than moving it to where we think it belongs.
An honest “we don’t know yet” is worth more than a confident misfiling. Anything in this category is named, sized, and dated on Where the Money Went, and it stays visible until it is resolved rather than being quietly absorbed.
How We Classify
Four rules, applied in order.
- Figures come from our books, never from a model. Every number we publish is pulled from QuickBooks. The figures we had to retract in August 2026 were modeled — someone’s reasonable estimate of what an organization like ours probably spends. We do not publish estimates as if they were accounts.
- Classify by what the money did, not by who spent it. A cost is program if it delivered the mission to somebody. Staff time is classified by the work, which is why our operations staff are program.
- When a cost could sit in either category, it goes to Management & General. We would rather understate our program share than inflate it. Every borderline call is made against our own interest, deliberately.
- When we cannot tell, we say so. It goes to Under review with the amount and the open question stated, rather than being assigned a plausible home.
Rule three is the one that matters most, and it is the reason our published program share is a floor rather than a target.
What Our Numbers Actually Are
| Category | 2025 | 2026 so far |
|---|---|---|
| Program Services | $107,730 · 88% | $54,734 · 93% |
| Management & General | $11,326 · 9% | $4,206 · 7% |
| Under review | $3,570 · 3% | — |
| Fundraising | — | $0 |
| Total expenses | $122,626 | $58,941 |
Against the BBB thresholds — at least 65% of total expenses on program, no more than 35% of related contributions on fundraising — we clear both comfortably. We would rather you read the next section than that sentence.
What the Ratio Cannot Tell You
A high program share is not the same as a healthy organization, and ours is proof.
In 2025 we spent $122,626 and took in $32,599 — a deficit of $90,027, covered from reserves. Individual donations were $6,953 of that revenue. An evaluator looking only at our 88% would score us well and miss entirely that we are spending down.
The ratio makes it worse, not better. Because 88% of our spending is program, there is no overhead layer to trim our way out of the deficit. If the shape does not change, what gets smaller is the aid — the grants and the two staff — not the office, because there is no office. A funder who rewards a high program share and then asks us to cut costs is asking us to cut the mission.
The largest labor input we have appears nowhere in these categories. Two of our co-creative directors work full time and draw no salary. There is no line for them because there is no transaction to record. Every percentage on this page is calculated on money, so the table describes a smaller organization than the one that exists — two people instead of four, with two senior people missing entirely. Unpaid labor makes the ratio look better and the organization more fragile at the same time.
We also do not count the non-monetary mutual aid our staff deliver — the appointments, the service navigation, the checking in. That work is continuous and relational and has no clean unit. Any number we produced would be a fact about our record-keeping rather than about the care.
On Overhead
Some evaluators expect most spending to go to programs. We agree with the spirit: donations should support real work, not vanity or extraction.
But the overhead ratio has done real harm to organizations like ours. It treats the infrastructure that makes work safe, lawful, and durable as a moral failing, and it pushes small organizations to underinvest in exactly the things — accounting, insurance, governance — whose absence later becomes a scandal. Every correction on this website exists because we pay an accountant. That is the administration line doing its job.
Operations are not waste. Operations are access infrastructure.
Access requires infrastructure. Care work is not free. Sustainability is part of dignity. We aim for strong program investment, lean and transparent administration, and non-extractive fundraising — and we would rather be judged on whether the money reached people than on which column it sat in.
When We Get It Wrong
We have. Twice in 2026, publicly.
On 9 August 2026 we corrected our published 2025 expenses: we had reported $120,000 split 65% program, 25% administration, and 10% fundraising. The real total was $122,626 and the real program share was 88%. On 10 August 2026 we flagged roughly $22,000 of reported investment income as possibly reserves being moved rather than earned — still with our accountant. On 3 September 2026 we corrected the same retracted figures where they had survived on Frequently Requested Information.
When a published figure turns out to be wrong, we commit to:
- Correcting it in place, with a dated note saying what it used to say. Not a silent swap.
- Logging it in the Transparency Log, so the correction is findable by someone who never saw the original.
- Chasing the same figure to every page carrying it. The September correction happened because the August one did not.
- Saying a number is shaky while we are still checking, rather than leaving it standing because it reads well.
A transparency page that only shows the good ratios is marketing.
What Changes When We Become a Public Charity
We are in the process of terminating private foundation status and becoming a public charity. Two things on this page change when that takes effect.
This classification stops being voluntary. Form 990 carries a statement of functional expenses that requires exactly this three-way split, line by line. What we publish here becomes what we file. The rules above were written to survive that.
Where our money comes from starts to matter as much as where it goes. Public charity status depends on a public support test, and our revenue is currently dominated by investment income rather than by donations from many people. That is a real problem for us and we would rather name it here than have it discovered. The revenue tables show the shape.
This page will be rewritten when the reclassification is approved.
Who Is Responsible
- Our accountant maintains the books and the account structure these categories are drawn from.
- The Executive Director applies the classification rules above and prepares the published figures.
- The President is accountable to the Board for what gets published, and for issuing a correction when it is wrong.
- The Board reviews the annual figures before they are published.
The Treasurer seat is currently vacant and the President is covering those duties. We note it here rather than leaving it implied — see Board & Governance.
Review of This Policy
The Board reviews this policy at least annually, alongside the year’s figures and with our accountant. It is reviewed again whenever the classification of a material line changes, and when our public charity reclassification takes effect.
Accountability is not policing. Accountability is care.
Document ID: SF-GOV-EXP-2.0
Status: Draft — pending adoption by the Board of Directors
Supersedes: version 1.0, published 15 February 2026, which described a classification we do not use and a fundraising function we do not have
Figures current as of: 9 August 2026 (2026 column); FY2025 complete
Related: Where the Money Went · Annual Reports · Disclosures · BBB Standards Alignment · Charting Impact · Gift Acceptance Policy · Conflict of Interest Policy · Document Retention & Destruction Policy · Transparency Log · Accountability & Transparency