I have sat in enough nonprofit board meetings to have watched this pattern play out more times than I can count. A staff member proposes hiring a part-time data analyst to actually track whether a program is working, or investing in a modest training budget so caseworkers do not burn out and leave within eighteen months, and the proposal gets tabled because it will nudge the overhead ratio in the wrong direction before the next annual report goes out. I understand why boards behave this way. Donors and watchdog rating sites have spent decades treating the percentage of a nonprofit's budget spent on administration and fundraising as the single most important signal of whether an organization deserves support. I think that habit has done real, measurable damage to the sector's ability to help anyone, and it is long past time we retired it.
Where the Overhead Obsession Comes From
The overhead ratio obsession comes from an understandable place. Donors want assurance that their money is going toward the mission rather than lining someone's pocket or funding a lavish office, and a low administrative percentage feels like an easy, legible proxy for integrity. I do not think that instinct is dishonest or foolish. Financial stewardship matters, and there absolutely are organizations that abuse donor trust with excessive executive compensation or wasteful spending, and donors are right to want protection against that.
A Proxy That Punishes Good Investment
The trouble is that overhead ratio, as commonly calculated and publicized, is a terrible proxy for the thing donors actually care about, which is impact. A nonprofit running on a starvation diet of administrative capacity often cannot afford basic infrastructure: a functioning database to track outcomes, adequate training for frontline staff, competitive enough salaries to retain experienced people instead of losing them to burnout every two years, or even proper financial controls, which is more than a little ironic given that the metric is supposed to signal financial trustworthiness. I have watched capable organizations chase an artificially low overhead number by understaffing case management, and the actual, measurable result was worse outcomes for the people the program existed to serve, not better ones.
There is also a fundraising perversity built into the whole framework. An organization that invests seriously in its fundraising operation, hiring skilled development staff, running proper donor stewardship, can often raise several dollars for every dollar spent on that function, dramatically expanding its total capacity to help people. But because fundraising expense counts against the overhead ratio regardless of how much revenue it generates, that same smart investment gets punished by watchdog scores as if it were waste. A rational organization responds to that incentive by underinvesting in fundraising capacity, which shrinks its total budget and its total impact, all in service of a number that was never actually measuring impact in the first place.
Steelmanning the Watchdog Argument
I want to steelman the defenders of overhead scrutiny here, because their underlying worry is legitimate even if their preferred metric is flawed. They point out that impact measurement is genuinely hard, expensive, and easy to manipulate, whereas an overhead percentage is at least simple, comparable across organizations, and difficult to fake outright. That is a fair practical concern. Rigorous outcome measurement takes real expertise and real money, which smaller organizations in particular may struggle to afford, and I do not want to wave away the difficulty of building good alternatives just because the current metric is flawed.
Measuring What Actually Matters
But difficulty is not a reason to keep using a measure that actively rewards organizations for underinvesting in their own effectiveness. The better path, and one that several thoughtful funders and rating organizations have already started moving toward, is requiring nonprofits to report specific, program-level outcome data: how many people were served, what changed for them, and how that compares to the cost of achieving it, evaluated over a multi-year horizon rather than a single fiscal year snapshot. That is harder to standardize than a simple percentage, and it requires donors to do more homework than checking a single score on a website. But it actually measures the thing donors say they care about, which is whether the organization is helping people, rather than a bookkeeping ratio that tells you almost nothing about that question and actively incentivizes organizations to starve the very capacity they need to succeed.
If we want a nonprofit sector capable of solving hard problems, we have to stop punishing it for building the infrastructure that hard problems require. Judge nonprofits by whether they help people. Stop judging them by how cheaply they manage to do it on paper.
