What question did your team argue about last meeting?

Before you build anything, ask the team what they argued about in their last meeting. Not what they want to see. Not what looks important. What did they actually disagree about, in real time, with stakes attached?

That’s your dashboard.

The 14-tab spreadsheet someone inherited in 2019 is not your dashboard. The template a consultant sold you in 2022 is not your dashboard. The funder report you cut and paste every quarter is not your dashboard. Those are artifacts. They describe activity. They don’t resolve disagreement.

A useful diagnostic: if the answer to what would change if this number went up or down is nothing, you’re tracking something nobody is using to decide anything.

Reports describe. Answers decide.

A good output ends with, “So we should…”, not “As you can see…”.

That single sentence is the cleanest way I know to test whether a report is doing its job. If you finish reading and you’re informed but not pointed at a choice, you’ve read a description. Descriptions are useful, but they’re a poor substitute for an answer.

Compare the two:

• Descriptive: Program attendance was 412 this quarter, up 8% from last quarter.

• Answering: Program attendance is up 8%, driven entirely by the Tuesday evening slot. Saturday morning has dropped three quarters in a row. We should either fix Saturday or move that capacity to Tuesday.

The data underneath both is identical. The work the report is doing is completely different. The first one tells you what happened. The second one tells you what to do about it.

If your monthly report is full of the first kind and nothing of the second, that’s why nobody opens it.

Why nonprofits build reports nobody reads

Reporting in nonprofits often gets built for funders first and staff second. That ordering quietly distorts everything downstream.

Funders need justification: proof that money was spent on the mission, proof that activity happened, proof that outcomes were measured. Those documents are designed to defend. They’re written to a specific audience with a specific question, and the question is usually some version of did you do what you said you’d do.

That’s a fine document. It’s just not the same document as one designed to help your program director decide whether to keep running the Saturday session.

When the funder document becomes the internal document by default, the people who could act on the data end up reading a report engineered to justify, not to inform. They skim it, file it, and make decisions the old way: by gut, by what was loudest in the last meeting, by what the longest-tenured person remembers from 2018.

The data is there. It’s just pointed at the wrong audience.

How do you replace the monthly report?

Stop thinking in documents. Start thinking in questions.

Replace your monthly report with a running list of live questions the team is actually trying to answer. Each question gets one view. When the question is answered or stops mattering, the view retires.

Here’s what that looks like in practice:

1. Capture the questions. At the end of every leadership meeting, write down what the team disagreed about or wasn’t sure about. Those go on the list.

2. Rank by stakes. Which question, if answered, would change a decision in the next 30 days? Start there.

3. Build one view per question. Not a dashboard with 12 charts. One view, one question, one clear read.

4. Set an expiration date. Every view gets reviewed quarterly. If nobody has referenced it, retire it. If the question has been settled, retire it.

5. Add new questions as they surface. The list is alive. It grows and shrinks with the work.

This sounds simple because it is. The difficult part isn’t the building. It’s letting yourself stop tracking things that aren’t tied to a decision.

What changes when you stop building reports nobody reads?

Three things, in roughly this order.

First, you build fewer charts. A lot fewer. The 14 tabs collapse into four or five views, because most of what was on those tabs was never tied to a choice anyone was making.

Second, the meetings get shorter. When the data points at a decision, the conversation moves from let’s review the numbers to let’s pick the option. That’s a different meeting, and it ends an hour earlier.

Third, and this is the quiet one: you notice how much of your tracking was performative. Tracking that existed because tracking felt responsible. Tracking that nobody used, nobody questioned, and nobody would have missed if it disappeared. Letting go of it feels uncomfortable at first, the way clearing out a closet always does. Then it feels like relief.

The mission gets more of your attention. The team makes faster decisions. The data finally earns its keep.

That’s the whole shift. Build for the question, not the audit. Retire what isn’t being used. Trust that fewer, sharper outputs will move the work forward more than a comprehensive report nobody reads.

Frequently asked questions

How is this different from a regular dashboard?

A regular dashboard shows metrics. A question-led dashboard shows the answer to a specific decision the team is trying to make. The difference is purpose: one is built to be comprehensive, the other is built to be useful. Comprehensive isn’t a substitute for useful.

What do we do about funder reporting?

Keep it. Funder reports serve a real purpose and have a real audience. The point isn’t to stop producing them. The point is to stop assuming they’re also the right document for internal decisions. They’re two different jobs that deserve two different outputs.

How many questions should be on the live list?

Fewer than you think. Most small nonprofits can run on five to eight active questions at a time. If the list grows past ten, you’re probably mixing strategic questions with operational noise. Split them, or cut the noise.

What if our team doesn’t know what they’re arguing about?

That’s its own diagnostic. If the team can’t name a disagreement or an open question, either the work is on autopilot (worth examining) or the meetings aren’t surfacing the real decisions (also worth examining). Either way, that’s the more important conversation to have first.

Do we need new software for this?

Almost never. Most teams already have the data and the tools. What’s missing is the question. A spreadsheet pointed at the right question beats a dashboard pointed at nothing.

If this kind of thinking is useful, the Meaningful Insights SubStack goes deeper into how small nonprofits can turn data into decisions without hiring an analytics team. Subscribe to for more notes on what to track, what to retire, and how to make your data earn its keep.

Subscribe now

Most teams skip straight to weighing options, gathering more data, or polling the room. They focus on collecting information. The real skill is knowing which question cuts through all of it. One well-formed question will save you from a dozen spreadsheets and three months of regret. This post is about that question, why it keeps getting skipped, and how to use it before your next big decision.

Why does the question nobody asks before making a decision get skipped?

Because it feels slower. And in most organisations, decisiveness is rewarded more than accuracy.

When a board is waiting, a grant deadline is closing, or a staff member is asking what to do next, asking “what would have to be true for this to be right?” feels like stalling. It feels like the opposite of leadership.

It isn’t. It’s the difference between a decision you can defend in twelve months and one you’ll quietly walk back.

The other reason it gets skipped: it surfaces uncertainty. Most decision-makers would rather feel confident than be correct. Naming the conditions that need to hold true forces you to admit which ones you can’t verify yet. That’s uncomfortable. It’s also where every good decision starts.

What’s wrong with gathering more information?

Nothing, until it becomes a substitute for thinking.

A pattern that comes up frequently in non-profit decision-making: “I have data, but I don’t know how to use it.” The instinct, almost always, is to gather more. More survey responses. More dashboards. Another report. The belief is that clarity lives somewhere on the other side of one more data pull.

It rarely does.

More information without a sharper question produces more noise, not more signal. You end up with three reports that disagree, no framework for weighing them, and a decision that defaults to whoever is loudest in the room. The question is the filter. Without it, the data has nowhere to land.

How does this question actually work in practice?

It works by inverting the usual flow. Instead of asking “what does the data say?” you ask “what would the data need to say for option A to be the right choice?”

Then you check.

Here’s a small example. A non-profit is deciding whether to open a second location. The board is split. Someone pulls foot-traffic estimates, someone else pulls donor maps, a third person makes the case from anecdotes. Nobody agrees.

Now apply the question. What would have to be true for a second location to be the right call?

• Demand in the new area would need to exceed unmet demand at the current location.

• The new site couldn’t cannibalise donors or volunteers from the existing one.

• Operating costs would need to stay within a defined percentage of program spend.

• There would need to be at least one local partner willing to share infrastructure.

Now the data has a job. Each item on that list is testable. You’re no longer drowning in reports. You’re answering four specific questions. And if even two of them come back negative, you have your answer, without another six months of debate.

When should you ask it?

Before you start gathering data. Not after.

This is the part most people get backwards. They collect first, then look for patterns, then try to retrofit a decision. By that point, sunk cost has already shaped the answer. You’ve spent three weeks on the analysis. Somebody is going to act on it, even if the conclusion is shaky.

Asking the question first does two useful things. It tells you what data you actually need (which is almost always less than you think). And it tells you what data is irrelevant, which protects your time and your team’s attention.

If you can’t articulate what would change your mind, you’re not ready to gather information yet. You’re ready to think.

What does this look like for a small non-profit?

It looks like fewer decisions made on anecdote, and more decisions made on conditions you can actually check.

Most small non-profits don’t lack data. They lack a way to interrogate it. Donation records, program attendance, volunteer hours, intake forms, social engagement, all of it sits in some system somewhere. The problem isn’t volume. It’s that nobody has asked a sharp enough question to make the data answerable.

Try this the next time a real decision is on the table:

1. Write down the decision in one sentence.

2. Write down what would have to be true for the answer to be yes.

3. Write down what would have to be true for the answer to be no.

4. For each item, mark whether you can check it with data you already have, data you could get, or data you’ll never have.

5. Start with the cheapest checks first.

You’ll often find the decision answers itself somewhere between step three and step five. Not because the data is magical, but because the question finally was.

What we keep getting wrong about decision-making

We treat decisions as moments. They’re not. They’re the visible end of a thinking process that started much earlier, with a question.

The quality of your decisions is almost entirely set by the quality of the question you asked at the beginning. Better data won’t rescue a vague question. More meetings won’t either. Senior advisors won’t, though they may help you ask a sharper one.

The organizations that punch above their weight, the ones doing more with less and serving their communities in meaningful ways, aren’t the ones with the most information. They’re the ones who have learned to ask the question nobody asks before making a decision, and to keep asking it, even when the room is impatient for an answer.

That habit is small. It costs nothing. It compounds for years.

Frequently asked questions

What if I genuinely don’t know what would have to be true?

That’s useful information. It means the decision isn’t ready to be made, or that you don’t yet understand the problem well enough to act on it. Sit with the question for a day, talk it through with one trusted person, and write down what comes up. The answer doesn’t have to be perfect. It just has to be specific enough to test.

How is this different from a pros and cons list?

A pros and cons list catalogs what you already think. This question forces you to define what would change your mind. That’s the key shift. You’re not measuring sentiment, you’re naming testable conditions. One produces opinions, the other produces decisions you can defend.

Doesn’t this slow everything down?

It feels slower for the first ten minutes and saves you weeks later. Most decision delays come from circling the same debate without resolution, not from thinking carefully up front. Asking the question early replaces a long, frustrating debate with a short, specific one.

Can this work for small, everyday decisions too?

Yes, though it’s overkill for most of them. Save it for decisions where the cost of being wrong is meaningful: hiring, program design, technology investments, partnership commitments. For those, a five-minute version of this exercise is worth more than another committee meeting.