Signs a nonprofit board has drifted

Nobody notices the meeting where a board stops governing. There is no vote on it. Attendance holds up, the minutes still look like minutes, everyone is still fond of each other, and the organization keeps running. Drift shows up as an absence, in the questions that used to get asked and don't anymore, and absences are hard to see from inside the room.

What follows is a set of symptoms you can actually observe, along with the check that tells you whether you have it. None of this is a judgment about anyone's commitment. Most drifted boards are made up of people doing their honest best inside a structure that quietly stopped asking anything of them.

The consent agenda swallows real decisions

A consent agenda is a good tool. It moves the minutes, the routine reports, and the uncontested renewals into one motion so the meeting can spend its time on things that need a conversation.

Drift is when the boundary moves. A vendor contract goes on consent because it is a renewal. A policy revision goes on consent because the committee already reviewed it. A budget amendment goes on consent because it is small. Each move is individually defensible and the cumulative effect is that the board approves a page of items it has not discussed.

The check: pull your last four consent agendas. For each item, ask whether the board would have been responsible if it had gone badly. Anything where the answer is yes does not belong on consent, no matter how uncontroversial it was.

The financial report is received, not read

The packet arrives, the treasurer walks through it, someone moves to accept, and it passes. This happens at thousands of meetings a month, and at a good fraction of them not one person in the room could tell you what the numbers mean.

It is almost never a competence problem. It is a format problem. Most board financial packets are the accounting system's default reports, which are built for people who already know the organization's chart of accounts.

The check: ask three board members, separately and casually, how many months of operating expenses the organization has in the bank. If you get three different answers, or three apologetic shrugs, the report is being received rather than read. The fix is usually one page with three numbers on it, cash on hand in months, year-to-date against budget, and the one variance that changed since last meeting, with the full statements attached behind it for anyone who wants them.

The same two people talk

Count it. Not impressionistically, actually count. Sit with the agenda and make a tally mark every time someone speaks for more than about 15 seconds.

On a drifted board, the chair, the chief executive, and one other person account for most of the talking, and the pattern has been stable for years. The rest of the room is not disengaged in any way they would recognize. They have simply learned that the decisions get made by those three, and that speaking up extends a meeting they are attending after a full day of work.

The sharper version of this signal is what happens to new members. A new board member asks good outsider questions for their first two or three meetings, then stops. That stopping point is the moment the board taught them the norm, and it is worth knowing what happened in those meetings.

The check: the tally, once. It takes no preparation and it is uncomfortable in a useful way.

Recruitment is whoever somebody already knows

Ask the board to name the last member who arrived from outside an existing member's personal network. If nobody can, your recruitment process is a friendship graph, and the board will keep converging on people who already agree with each other about what the organization is for.

The related symptoms travel together. Seats get filled reactively when someone resigns rather than against a list of what the board is missing. Term limits exist in the bylaws and are not applied, because the treasurer is the only one who understands the finances and losing him is unthinkable. The nominating committee meets once a year, in a hurry, in May.

BoardSource's periodic surveys of nonprofit boards have found for years that boards rate themselves lowest on exactly this cluster: recruitment, fundraising, and assessing their own performance. It is a sector-wide pattern rather than a failing specific to yours.

The check: write down what the board would need if two people left tomorrow. If the answer is "two more people," the board has no recruitment strategy, just a headcount.

Meetings end early because there is nothing to decide

This one gets mistaken for efficiency. The board meets for 50 minutes of a scheduled 90, everyone is pleased, and the reason is that the meeting consisted of the chief executive reporting and the board receiving.

A board that only receives is an audience. It is a comfortable arrangement and it usually suits everybody in the short run, including the chief executive, who has more than enough to do without a room of part-time volunteers reopening settled questions.

The check: read the last 12 months of minutes in one sitting and count the actual decisions. Not approvals of things already done, not acceptances of reports. Decisions where more than one course of action was genuinely on the table. If the count for a year is under five, you have a reporting body rather than a governing one.

Three other things surface in that same read. Whether anyone has looked at the bylaws in the past five years. Whether the chief executive has had a real performance conversation, in writing, recently. And whether the same fundraising discussion appears three times with no target attached to it.

What to check first

If you only do one thing, do the minutes read. An hour with 12 months of minutes is the fastest diagnostic in board work, it costs nothing, and it is very hard to argue with the results because the board wrote them.

After that, in order:

• Ask every member, individually, what they think the board's main job is over the next 12 months. Compare the answers. A board where six people give six unrelated answers has a purpose problem, not a process problem.
• Count who talks.
• Check whether any board member could explain the organization's cash position without opening the packet.

The awkward part of self-diagnosis is that the people doing it are the same people who drifted, and everyone in the room has social reasons to be generous about it. That is why the useful version of this asks everyone at once, separately, without names attached, and then compares the answers to each other.

That is the whole design of this assessment. The board questions cover the specific practices above, the chief executive answers the same questions with their responses reported in their own column rather than averaged into the board's, and the gap between those two columns is usually the most informative thing on the page. Asking the staff the same kind of questions tends to sharpen it further, because staff see the consequences of board drift before the board does. It is free, it takes about eight minutes per person, and no answer is ever linked back to the person who gave it.

Two honest limits. Small groups get sections withheld rather than shown: board results need at least 4 responses and 60% of the roster before a section appears, and there is no way around that, because a section built on two answers tells you about two people. And an assessment does not fix anything. It shows a board what it looks like from several angles at once. What happens after that is the board's work, which is why the cadence you set for repeating it matters more than the first result.

Drift is not decline. Most of the boards this describes are attached to organizations doing good work, staffed by people who show up. The slide is structural: meetings optimized for smoothness, agendas optimized for finishing, recruitment optimized for ease. Each of those is a reasonable local choice, and together they produce a board with nothing left to decide. The good news in that is that the same mechanism runs in reverse, one agenda at a time.