Nonprofit board committees: how many do you need?

A seven-person board with four standing committees is asking each member to serve on two of them, chair one, and still read the packet. That is not a structure, it is a wish. Nonprofit board committees get created the way most board structures get created, which is by copying a template from a governance article or from the last organization someone served on, and then never being reviewed again. The committees that resulted may be doing real work. Some of them almost certainly stopped years ago and nobody has said it out loud.

The honest answer to how many you need is usually fewer than you have, and for a board under about nine people it may be one.

The standing committees most boards have

Three show up almost everywhere, and they are not equally load-bearing.

Finance, sometimes audit. Reviews the financials before the full board sees them, works with the treasurer on the budget, and asks the questions that are hard to ask in a room of 10 people in 12 minutes. If your organization has a real audit, an audit committee separate from finance is a genuine control, because the people overseeing the audit should not be the people who prepared what is being audited. Below a couple of million dollars in budget, most organizations combine the two and say so.

Governance, sometimes nominating. Owns recruitment, orientation, the board's own composition, term tracking, and the assessment cycle. A nominating committee that only wakes up when a seat opens is doing a fraction of the job. This is the committee most likely to matter and most likely not to exist.

Executive. Usually the officers plus the immediate past chair, authorized by the bylaws to act between meetings. This one deserves its own section below, because it is the one that goes wrong quietly.

Beyond those, boards commonly add development, program, and personnel committees. Each can be justified. Each also costs meetings, minutes, a chair, and a slot on the agenda, and the cost lands on volunteers who already have jobs.

The three-part test for nonprofit board committees

Before adding one, and once a year for the ones you have, ask three questions. A committee that fails any of them is not a committee.

Does it have a written charter? One page. What it is responsible for, what it decides versus what it recommends, who is on it, how often it meets, and what it reports to the board. If nobody can produce this document, the committee's scope is whatever its chair believes it is, which is how two committees end up doing the same work and how a third does none. The instrument asks board members directly whether "Our committees have clear charters describing what they are responsible for."

Does it actually meet? Not "is scheduled to meet." Count the meetings in the past 12 months and compare that to the plan. A committee that met twice against a plan of six is telling you something true about how much the work is needed, and the honest response is to change the plan or close the committee, not to feel guilty.

Does its work change a board decision? This is the one that separates a working committee from a ceremonial one. A committee that produces a report the board receives and files has not changed anything. A committee that brings two options, a recommendation, and a reason is doing the work that justifies the meetings. The matching item is worth reading twice: "Committee work meaningfully shapes full board decisions, not just reports to it."

Run the test on each of your committees this quarter. Most boards find one that passes all three, one that passes two, and one that has not passed any of them since 2023.

What a small board genuinely needs

For a board of five to nine people, the defensible minimum is one standing committee, and it is finance or audit.

The reasoning is that financial oversight needs a smaller group looking closely and regularly at something the full board sees monthly and briefly. Everything else a small board does can happen in the full meeting, because the full meeting is seven people, which is roughly the size a committee would be anyway. Splitting seven people into three committees produces three meetings of two or three people plus the meeting they were all going to attend.

Governance work still has to happen. On a small board it usually belongs to the chair plus one other member, working as a named pair rather than a committee with minutes. Call it what it is. That is not a downgrade, and pretending it is a standing committee just adds a report to an agenda that already has too many. The same logic runs through everything about running a board self-assessment on a small board, where the structures that work at 18 members mostly do not survive the trip down to seven.

If your board is larger, the arithmetic changes. Around 12 to 15 members you have enough people that committees are the only way to give everyone real work, and the question shifts from whether to have them to whether each still passes the test. Board size and committee count move together, which is one more reason to be deliberate about how many board members you actually need.

Task forces with an end date are the better default

The better structure for most small boards is not a standing committee at all. It is a task force with a specific question, three or four people, and a date it stops existing.

Rewrite the employee handbook by March. Choose a new database by the June meeting. Plan the 20th anniversary event. Each has a finish line, and a group that knows it disbands in five months behaves differently from a committee that meets forever. People say yes more readily to a defined commitment. Progress is visible because the endpoint is defined. And nobody has to make the awkward motion to dissolve something, because dissolution was in the plan.

The failure to avoid is the task force that quietly becomes permanent. Write the end date in the motion that creates it. If the work is not finished, the board renews it deliberately, which takes one minute and forces someone to say the work is still worth doing.

The executive committee that becomes the real board

This is the failure mode worth watching for, because it happens without any decision to let it happen.

The bylaws let the executive committee act between meetings. That is sensible. Then a decision comes up two weeks before the meeting and the executive committee handles it. Then another. Within a year the officers are meeting every month, the substantive discussion happens there, and the full board meeting has become a briefing on what four people already decided. Nobody chose this. It just turned out to be faster.

The tell is on the other side of the table. Board members stop reading the packet, because the outcome is settled before the meeting and reading it changes nothing. Attendance drifts. The people who notice first are usually the newest members, who then get quiet. That pattern shows up alongside the other signs a board has drifted.

Three guardrails, none of them expensive. Limit the executive committee in the bylaws to genuinely time-sensitive matters and name the categories it cannot touch, which should include the budget, the chief executive's hiring or evaluation, and anything structural. Report every executive committee action to the full board in writing at the next meeting, as an item, not a footnote. And if the executive committee is meeting monthly, treat that as evidence the full board is not meeting often enough, rather than evidence the executive committee is efficient.

The IRS is interested in this too. Form 990 Part VI asks whether the governing body delegated broad authority to an executive committee, and the answer is public, sitting in the Form 990 that funders read.

If you want to know what your own board thinks of its committees rather than what the chair thinks, the assessment includes an optional set of questions on committee effectiveness covering charters, whether committees met as planned, and whether their work shapes decisions. Answers come back anonymously and are compared against similar organizations rather than judged on their own. It is free, and there is nothing to buy.

Pick your least defensible committee. Ask whether it has a charter, whether it met, and whether the board decided anything differently because of it. If the answer is no three times, closing it is not a loss. It is 12 hours a year back for people who did not have them.