What is a board self-assessment?

Someone at your last meeting asked what a board self-assessment is actually for, and the best answer anyone had was that it's good governance practice. That answer gets a motion passed and leaves everyone slightly unconvinced, which is roughly the worst possible starting point for a survey that depends on people answering candidly.

So, plainly. A board self-assessment is a fixed set of questions that every board member answers privately about how the board is doing its job. The answers are pooled, no individual's responses are visible to anyone, and the board reads the totals together. That is the entire mechanism. Everything that makes one useful or useless comes down to what you ask, who else you ask, and what happens in the meeting afterward.

What a self-assessment actually measures

It measures the board's own work. Not the organization's program outcomes, not the executive's performance, and not any individual member's contribution.

That work gets broken into domains. The board questions cover mission and strategic direction, financial oversight, executive support and evaluation, fundraising, composition and recruitment, culture and engagement, meetings and information flow, the line between governance and management, and governance practices. Each domain is a small set of statements a member rates on an agreement scale.

The statements are deliberately concrete. "The board receives financial information early enough and in a clear enough format to ask real questions" is a different question from "we have good financial oversight," and only one of them is answerable. "I can raise bad news with this board without it damaging me" tells you something about culture that no attendance record will.

Alongside the rated statements sit factual ones, answered yes, no, or don't know. "The chief executive received a documented performance review in the past 12 months" has a real answer. What makes it interesting is when the board disagrees about it. If four members say yes, three say no, and two say they don't know, the finding is not really about the review. It's about how little the board knows about its own recent history.

Don't know is treated as an answer, not a gap. If a third of a board doesn't know how the organization's finances work, that is a result worth reading out loud.

What it is not

It is not an evaluation of the chief executive. These two things get conflated constantly, partly because both involve a form and a difficult conversation. A self-assessment asks whether the board has a process for evaluating the executive, whether it used that process in the last year, and whether it supports the executive as well as it oversees them. It does not rate the executive. If you want that, run it separately, on its own schedule, with its own questions.

It is not a performance review of individual board members. Nobody gets a score. There is no ranking, no per-member column, and no way to see how any one person answered. A board that wants to address one member's attendance has to do that the hard way, in a conversation, the same as always.

It is not a compliance exercise. Form 990 Part VI asks whether you have a conflict of interest policy, a whistleblower policy, and a document retention policy, and whether the full board saw the return before it was filed. Those questions come from the IRS, and answering them yes is a matter of record-keeping. A self-assessment overlaps with that territory and then keeps going, because a policy existing and a policy functioning are different facts. Plenty of boards can point to a conflict of interest policy that no current member has read.

It is not a satisfaction survey. Whether board members enjoy meetings is a real question, and it's a small part of a much larger one.

What a board gets out of one

Three things, roughly in order of value.

The first is a shared picture where there were previously nine private ones. Board members form opinions about how the board is doing and mostly keep them to themselves, because saying "I don't think we understand the finances" out loud in a meeting sounds like an accusation aimed at the treasurer. Aggregated, the same sentence is just information.

The second is divergence, and this is usually the most useful output. When you ask the chief executive the same questions and report their answers as a separate column rather than folding them into the board average, the gaps become visible. Imagine an invented nine-member board where seven members agree that the board gives the executive meaningful support, and the executive disagrees. Neither number is interesting alone. Together they describe a specific problem and point at the conversation that resolves it. The same holds for staff and for the people you serve, which is why staff and community questions exist alongside the board set.

The third is a baseline. One round tells you where you stand relative to comparable organizations, and not much about yourself. Two rounds tell you which direction you're moving. This is the main argument for treating an assessment as an annual habit rather than an occasional project, which is covered in more depth in how often a board should evaluate itself.

One caution on reading results. A domain average means very little without something to compare it to. A 3.6 on fundraising is not low, or high, until you know what similar organizations score, and if there aren't enough comparable organizations to say, the honest report says that rather than implying a verdict.

Before you run one

Four things worth settling before anyone sees a question.

Decide who you're asking. The board alone is a room grading its own homework. Adding the chief executive costs nothing and roughly doubles what you learn.

Decide what happens with the results, and say so. "We'll spend 45 minutes on this at the March meeting" is a commitment people will answer honestly for. "We'll see what comes back" is not.

Tell people the confidentiality terms before question one, in specific terms. Not "responses are anonymous," but what is actually stored and what your board will see. People calibrate their candor to the mechanism, not to the adjective.

Accept that the first round will be generous. People are kind the first time they're asked. The signal shows up in round two.

If you want the answers rather than just the questions, this assessment collects them anonymously and shows the board and chief executive views side by side. It takes about eight minutes per board member, it's free, and there's no version of it that isn't.

A self-assessment doesn't fix a board. It shows a board what it looks like from the inside, in enough detail that the next conversation can be about something specific. What that conversation is, and whether anyone acts on it, was always going to be up to you.