The board report that answers no questions

Best fit: board governance, nonprofit leadership, executive director audiences

Every quarter, a development director spends the better part of two days producing a board report. Every quarter, a board member asks a question the report does not answer, and the honest response is that it will take a week to find out.

The report is not bad. It is usually accurate, well presented and delivered on time. It is just built to be read rather than to be interrogated, and boards interrogate.

The shape of the problem

A board report is a snapshot. Totals, comparisons to last year, progress against goal, a few highlights. It is designed as a document, which means every number in it is a dead end. You can read it. You cannot ask it anything.

The questions boards actually ask are almost all follow-ups. Which segment drove that increase. Are the lapsed donors from last year the same people as the year before. How much of that total is one gift. What would the number look like without the campaign. Is the retention drop concentrated in a giving level or spread across the file.

Each of those is answerable in principle. In practice, each one requires someone to go back to the database, build a query, check it, and produce a follow-up memo that arrives after the meeting has moved on.

So the questions stop being asked. Not because boards lose interest, but because they learn that questions are expensive, and a board that has learned not to ask questions is not providing oversight.

Three fixes that do not require new software

Report the question, not just the number. For each figure in the report, state what it does and does not include. "Total raised" that silently includes a bequest, or excludes pledges, will generate a question every single time until someone writes down the definition. This alone eliminates a surprising share of follow-ups.

Anticipate the three obvious ones. Before the meeting, look at each headline number and ask what you would want to know next if you saw it cold. Put those three answers in an appendix. It is twenty minutes of work and it converts the meeting from data clarification into discussion.

Fix the definitions once. Most follow-up questions in this category are definitional rather than analytical. Which formula produced the cost of fundraising ratio, whether retention is counted by donor or by dollar, what counts as a major gift this year. Agreeing these in writing once removes the same conversation from every future meeting. We went through where the standard definitions come from, and where they conflict, in a guide to fundraising metrics and benchmarks.

Why it stays broken

The underlying reason is that the data required to answer follow-ups is not in one place.

Giving history is in the donor database. Campaign attribution may be in an email platform. Event attendance is in a spreadsheet or a ticketing tool. Grant income is tracked separately. The relationship context that would explain why a segment moved is in notes fields, old proposals and people's heads.

Answering "why did major gifts drop this quarter" requires joining at least three of those, which is why it takes a week. It is not that nobody thought of the question. It is that the question crosses systems, and crossing systems is a project.

This is the actual case for nonprofit reporting dashboards, and it is a different case than the one usually made. The value is not prettier charts. It is that a question asked in a meeting can be answered in the meeting, which changes what a board meeting is for.

What good looks like

A useful test: in your last four board meetings, how many questions were answered in the room, and how many became action items?

If the ratio is heavily toward action items, the reporting is functioning as a broadcast rather than a conversation. That is a fixable problem, and the first fix is free. Write down what each number means.

The second fix costs something, and it is worth being honest that it is a real project. But the return is not efficiency. It is that your board can do the job it exists to do, which requires being able to ask a second question.