Most owner-managed businesses get a monthly or quarterly pack from their accountant, read it with mild dissatisfaction, and file it. The numbers are correct. The presentation is professional. And somehow it doesn't tell them what to do.
That isn't a criticism of accountants. It's a mismatch of purpose. Management accounts are built for one job and running a business next week is a different one — and knowing the difference tells you what to ask for instead of grumbling about what you get.
What management accounts are actually for
They exist to establish, accurately and consistently, what happened financially in a completed period. That requires a set of properties that pull directly against usefulness for operational decisions:
They're periodic. Monthly or quarterly, because that's the unit accounting works in and the unit that reconciles. Businesses don't operate in months; they operate in weeks and days.
They're retrospective and lagged. They arrive after the period closes, and closing takes time. By the time you read the picture, it's three to six weeks old.
They're financial. Revenue, cost, margin, balance sheet. They are structurally silent on the things that cause those numbers: which customers are drifting, how much work is committed, whether you have capacity next month.
They're aggregated. Categories exist for accounting comparability, not for operational relevance. "Cost of sales" is one line whether the interesting story is one supplier's price rise or a change in mix.
Every one of those properties is correct for the job they're doing. They also mean the pack can't tell you what to do on Tuesday.
Management accounts tell you what happened. Management information tells you what to do about it.
You need both, they're different documents, and only one of them is your accountant's job.
The four questions accounts structurally can't answer
Which customers are going quiet?
Accounts show revenue by period, not behaviour by customer. A month where three good customers stopped ordering and two new ones started can look identical to a flat, unremarkable month. The information needed — each customer's ordering rhythm compared with their own history — exists in your order data and doesn't appear in the accounts at all.
What's coming?
Accounts are, by construction, about what has been invoiced. The order book, the pipeline, the forward bookings, the work in progress — the things that tell you about next quarter — are outside their scope. A business can have a superb month and an empty order book, and the pack will look excellent.
Have we got the capacity?
Staff cost appears as a number. Whether that staff has eleven days of scheduled work in a five-day week does not. For any business selling time or production capacity, that's the constraint that actually determines next month's revenue.
Why did it move?
A variance shows that gross margin fell two points. Establishing whether that's mix, discounting, a supplier increase or one bad job requires cutting the data in ways the accounts don't hold — usually joining financial data to operational data that lives in a different system.
What management information looks like instead
Different cadence, different content, different point.
Weekly, not monthly. Frequent enough that intervention is still possible. A problem found in week two is a conversation; the same problem found in the following month's pack is a loss.
Operational and financial together. Sales alongside the order book. Margin alongside the jobs that produced it. Cash alongside who owes it and how old it is.
Comparison, not totals. Against the recent average and against the same week last year. A number without a comparison is trivia.
Exceptions, not just aggregates. The specific thing that has slipped — that account, that quote, that job — because that's what someone can actually act on.
Interpretation included. One sentence explaining why each number moved. This is the part that converts a report from a scoreboard into a decision.
Why this normally doesn't get built
Because it spans systems, and nobody owns the space between them.
Your accountant owns the financial system and can produce anything within it beautifully. Your operational software owns jobs, or stock, or bookings. Neither can see across, so the joined-up view falls to whoever in the business is willing to spend a Friday afternoon in a spreadsheet — which is why it happens sporadically and stops whenever things get busy.
That's the actual gap. It's not a skills gap or a software gap. It's that the useful information lives in the space between two systems that nobody is responsible for.
What to ask your accountant for
Two things, both entirely reasonable and both often refused only because nobody asked.
Programmatic access to the ledger. Read-only API access to your own accounting data, so it can be joined to your operational data. It's your data. There's no good reason for it to be reachable only through their pack.
The definitions, written down. What exactly is in "sales", what's in "cost of sales", how revenue is recognised, how work in progress is treated. That document is the thing that lets any other reporting agree with the accounts — which is what stops the whole exercise turning into an argument about whose number is right.
What a weekly brief contains that a pack doesn't
Concretely, side by side, for the same business in the same week.
| The pack says | The brief says |
|---|---|
| Revenue £412,000, up 3% on prior month | Last week £41k, 9% above the four-week average, almost all of it one product line |
| Debtors £84,000 | £31k overdue, £22k of it one customer now 71 days out, and the total has worsened three weeks running |
| Gross margin 31.2%, down 0.8pts | Margin fell because two large jobs ran over on labour — both for the same client, both quoted on the old rate card |
| (silent) | Three accounts worth £2.1k a month have stopped ordering |
| (silent) | Order book covers six weeks; normal is nine |
| (silent) | Two quotes from three weeks ago have had no follow-up |
The left column is accurate and reconciled. The right column is what somebody can act on before Friday. Three of the six rows on the right are structurally invisible to the left, because they aren't financial facts about a closed period.
Read any line and ask: could someone do something about this in the next five working days?
If yes, it's management information. If it's a fact about a period that has already closed, it's an account. Both are necessary; only one of them changes what happens next week.
Why the lag is structural rather than lazy
It's worth understanding why the pack arrives when it does, because it stops the frustration being aimed at the wrong place.
A month can't close until purchase invoices have arrived, accruals and prepayments are calculated, bank and control accounts are reconciled, and stock is valued. Several of those depend on third parties — you cannot accrue for a supplier invoice that hasn't turned up.
So the lag isn't slowness, it's the price of accuracy. And that's exactly why asking your accountant to produce faster management accounts is the wrong request: you'd be asking them to trade away the property that makes the pack worth having.
The right request is different data, not the same data sooner. Operational reporting can be fast because it doesn't have to reconcile — an indicative sales figure that's 98% right on Monday is more useful for management than a perfect one in four weeks, provided nobody confuses it with the accounts.
Making the two agree
The obvious risk with weekly operational reporting is that it disagrees with the accounts, and then nobody trusts either.
Three things prevent it. Write the definitions down, so the weekly figure states plainly whether it's net of VAT and net of discount, and at order or invoice date. Reconcile once, properly, at the start — take a closed month, produce the same numbers both ways, and chase every difference until it's explained. And label clearly: the weekly brief is indicative and operational; the pack is the reconciled truth. When they differ, the pack wins and the difference should be explainable.
Do that and the two become complementary. Skip it and you've added an argument to every meeting.
What not to conclude
Not that the accounts are worthless. They're the reconciled truth, and any operational reporting that disagrees with them is wrong until proven otherwise. They're also what your bank, your investors and HMRC care about.
And not that your accountant should be building the other thing. Some will and some are very good at it, but it's a different discipline with a different cadence, and it needs read access to systems the accountant has no business being in.
What to conclude is narrower and more useful: the pack you're mildly dissatisfied with is doing its job correctly, the thing you actually want is a different document, and nobody is currently responsible for producing it.