Someone asks a simple question about your numbers. Why does answering it take three days?

The problem

The gap catches you twice: when someone outside asks, and when you have to decide.

The report goes out and it is fine. Then a funder, a lender, or your own board asks how a number was arrived at — or you need to know what's working right now, not next April — and the answer takes three weeks and four people.

It is rarely a software problem. Financial data sits in one system, delivery data in another, and the join between them is nobody's job — so it gets done by hand, by someone who is good at it, and the cost never appears in a budget because it is absorbed rather than spent. Here is what that looks like from inside. Most organizations will recognise two of these.

• You can produce the number, but not defend it

  If you run a nonprofit

  Tuesday, 4pm. Your largest funder's programme officer emails about the outcomes figure in your last report: "could you walk us through how that was arrived at? We're doing this across our portfolio." You forward it to the person who wrote the report. She forwards it to two people. What comes back is a narrative, not a trace — and everyone involved knows the difference, including the funder.

  If you run a company

  You're raising to fund next year's growth, renewing your credit line, or in early conversations to sell. The analyst asks for the detail behind gross margin by product line. Two weeks later your team sends a spreadsheet with a tab called final_v3_FIXED. Nothing in it is wrong. But the rate, or the price, quietly moves against you — because how long it took told them something the numbers didn't.

  credibility, at the only moment it is actually being measured. The report bought you nothing; the follow-up is where they decided what you are.

• The same question gets two different answers

  If you run a nonprofit

  Board meeting. Slide eight says you served 4,100 people this year. A trustee flips back to the programme report from March, which says 4,600 for the same period, and asks — not unkindly — which one is right. Both are. They count different things, for good reasons nobody ever wrote down. You spend ten minutes explaining your data instead of your strategy.

  If you run a company

  Monday review. Sales says the quarter closed at $2.3M. Finance says $2.1M. Both are right — one counts bookings, the other invoiced, and the difference has never mattered enough to settle. Forty minutes go to whose number is real. Zero go to why the quarter came in soft, which was the meeting's actual question.

  days of reconciliation before every deadline, and a room that quietly starts discounting every number you bring it.

• One person can stop your reporting

  If you run a nonprofit

  Your finance manager asks for a meeting and you feel it in your stomach before she says anything — because if it's a resignation, you're not losing an employee, you're losing the only person who knows how the spreadsheet actually works. Which tabs are stale. Which column to ignore. Why March is different.

  If you run a company

  Your controller has been with you eleven years. She's the only person alive who knows why the job-costing sheet and the accounting system never quite agree, and month-end doesn't close without her. She mentions, in passing, that her husband's company may relocate him. You hear it as an operational event, because it is one.

  a quarter you cannot report on, hinging on one person's life decisions. Every organization I look at carries this risk. Almost none have priced it.

• You are deciding on instinct

  If you run a nonprofit

  One of your sites has been slipping since February. You'll see it clearly in the annual figures — next year, reconciled and aggregated, when it's a fact instead of a signal. Right now it's a hunch you can't substantiate, so it loses the argument to whoever is most confident in the room.

  If you run a company

  You suspect one branch — or one crew, or one product line — is losing money on every job. You'll know for certain at year-end. In the meantime you're quoting new work at the same prices and hiring against the same plan, on a hunch that lost the argument last quarter because it was only a hunch.

  the decisions themselves. You are running the organization on judgement because the data that would sharpen it arrives after the moment to use it.

Reporting is the obligation. Deciding is the point. A number that satisfies an outside party and a number your management team can act on are not the same thing. The second is the harder one, and it is the one I hold the work to.

If you read one of those and thought of a specific email, a specific meeting, or a specific person — that's the one to tell me about. It takes one conversation to find out whether it's worth two weeks of looking, and I'll say plainly if it isn't.

Book a 30-minute call Or write to me first

When people call me

Almost always within a month of one of these.

The gap is always there. What changes is that somebody new acquires the right to ask about it.

• A government grant ended and you are replacing it with several foundation grants — five reporting regimes where there was one.
• You signed an outcome-contingent contract — value-based care, a Medicaid line, a placement-rate condition. Revenue now depends on numbers you have never had to defend.
• An acquisition closed and the two companies' KPIs do not reconcile, because each arrived with its own chart of accounts.
• A sale process or a credit-line renewal is starting and diligence is about to go through your numbers line by line.
• A new CFO arrived and is asking where each figure comes from — which nobody has been asked in years.

If one of those has happened, or is about to, the two weeks are worth spending now rather than during the scrutiny.

Book a 30-minute call

And if none of those apply

The gap is costing you every week nobody is asking about it.

Most organizations I speak to are not in a window. Nothing is due, nobody is examining anything, and the reporting goes out more or less on time. The cost is still being paid — it is just being paid quietly.

This is not a rare condition. Of a thousand grant professionals surveyed in 2026, only 18% could answer a basic budget question within minutes. 96% ran at least one workaround outside their official systems, 61% kept a personal spreadsheet for budget tracking, and 55% relied on knowledge held in one individual's memory. On the commercial side the median finance team takes eight days to close its books, and 73% of finance staff work overtime to get it done — about eleven extra hours per cycle. Three quarters of executives say they do not trust their own numbers.

Ask how many hours a single reporting cycle takes, and whose hours they are. Multiply by how often that cycle runs. Four hours a week is five working weeks a year at a senior salary — real money, spent proving work you have already done. Nobody has ever put that figure in front of your board, because nobody has ever added it up.

The second cost is slower and larger. When the numbers only reconcile in arrears, decisions get made on instinct — which programme to expand, which branch is drifting, which crew is actually profitable — while the data that would have answered it is still being assembled. That happens every quarter, trigger or no trigger.

And the trigger does come eventually. Every organization in this range meets one — a funder changes, a lender asks, someone wants to buy you, a new CFO starts. Being ready in advance costs a fraction of getting ready under a deadline, and it is the same two weeks either way.

Book a 30-minute call

Recent work

800 nonprofit filings, read end to end.

You don't have to take any of this on trust — the gap is measurable. Nonprofits are the one sector where it is measurable from outside, because their filings are public. So I read the public filings of 800 mid-sized American nonprofits to see whether their reported numbers trace back to anything. The pattern is not specific to nonprofits — it is just visible there.

I wrote up what it is, and what it costs →