Every trade business that's failed on my watch had the same final chapter. Not a lack of work — a Tuesday morning where wages were due, a big invoice hadn't landed, and the owner found out both facts at the same time.
Profit problems kill businesses slowly. Cash problems kill them on a specific date. And the difference between a cash problem that's survivable and one that isn't is almost always notice — how many weeks before the bad week you knew it was coming.
That's the entire job of a 13-week cash flow forecast. It's not an accounting document and it doesn't care about accruals or depreciation. It answers one question, thirteen times: what will actually be in the bank at the end of each of the next thirteen weeks?
You can build one in an afternoon with a spreadsheet. Here's how.
Why 13 weeks
Thirteen weeks is one quarter, and it's the sweet spot between two failures.
Shorter horizons — the "can we cover Friday" mental maths most owners run — give you no time to act. By the time a two-week view shows trouble, your options are down to overdraft or grovelling.
Longer horizons turn to fiction. You genuinely know a lot about the next quarter: the jobs on the books, the invoices outstanding, the BAS that's due, the team you're paying. Week 26 is guesswork wearing a spreadsheet costume.
Thirteen weeks captures every quarterly obligation at least once and gives you enough runway to fix what it reveals. A problem visible six weeks out has half a dozen solutions. The same problem visible on Monday morning has none.
The seven rows that matter
Open a spreadsheet. Thirteen columns, one per week, starting this week. Then these rows:
1. Opening balance. The actual bank balance at the start of the week. Week one is today's number; every other week inherits the closing balance of the week before.
2. Receipts — invoiced work. Every outstanding invoice, placed in the week you expect the money to arrive. More on this in a moment, because this row is where forecasts live or die.
3. Receipts — pipeline work. Jobs booked but not yet invoiced: deposits due, progress claims, completions. Date them by when the cash lands, not when the work happens.
4. Wages and super. Every pay run, in its week. Since 1 July, super leaves with wages — if Payday Super has changed the rhythm of your outgoings, this forecast is where the new rhythm becomes visible instead of surprising.
5. Materials and subbies. Tied to the jobs in row 3. If a job's revenue appears in week six, its materials usually appear in weeks four and five — that ordering, spend-before-receipt, is precisely the squeeze a forecast exists to show you.
6. The lumpy stuff. BAS. Insurance renewals. Vehicle registrations. Loan repayments. Equipment purchases. Tax instalments. Each in the specific week it leaves the account.
7. Closing balance. Opening balance plus receipts minus outgoings. This row is the forecast. Everything else is working.
Gold nugget. The lumpy stuff in row 6 is the row that kills trade businesses, so build it from evidence, not memory: pull up the last twelve months of bank statements and write down every single withdrawal over $2,000 that wasn't wages or materials. That list — with its dates — is your row 6 for the next year. Most owners who do this exercise find two or three five-figure annual payments they'd forgotten were coming.
Date receipts by when customers actually pay
Here's the mistake that makes most first forecasts useless: putting invoices in the week they're due.
Your terms might say 14 days. Your customers pay in the timeframe your customers pay in, and you already have the data to know what that is. Pull your last 20 or 30 paid invoices and calculate the average gap between issue date and money-in-bank. For most trade businesses it's somewhere between 25 and 45 days, regardless of what the invoice says.
That number — your real debtor days — is the lag you apply to every receipt in rows 2 and 3. An invoice issued this week with 35-day real terms is week-five money, and pretending otherwise doesn't make it arrive faster; it just makes your forecast lie to you in the direction of comfort.
Gold nugget. Segment the lag by customer type and the forecast gets sharply better: homeowners often pay in under a week, builders in 45 to 60 days, real estate agencies and property managers on their own monthly run whatever your terms say. Three lag numbers instead of one is usually the single biggest accuracy upgrade available — and it also tells you exactly which customers are financing their business with your money, which is a pricing conversation. If someone's genuinely not paying at all, that's a different playbook: what to do when a client doesn't pay.
Read it like an operator, not an accountant
With the rows filled in, the closing-balance line tells you one of three stories.
Every week comfortably positive: good — now raise the bar. Add a row for a minimum buffer (a month of wages-plus-super is a sensible floor) and treat any week that dips below the buffer, not zero, as the alarm.
A dip that recovers: the most common story, and the most useful one. You can see exactly which week pinches and exactly why — usually a BAS week colliding with a builder's 60-day payment. With six weeks' notice you can pull an invoice forward, chase two debtors, shift a materials order, delay a discretionary purchase, or arrange finance from a position of calm. All five of those options expire as the week gets closer.
A slide that doesn't recover: the forecast has just told you the truth most owners learn a year later — the business model itself is underwater, and no amount of timing tricks fixes a margin problem. That's not a forecasting issue; it's pricing or job selection, and the fix lives in how to price your services so you actually make money and in checking your real margins with the job margin calculator. The forecast's contribution is telling you now, while there's cash left to fund the fix.
The Friday habit that makes it real
A 13-week forecast built once is a photograph — accurate for a week, then decoration. The value is in the update rhythm, and the rhythm is fifteen minutes every Friday:
- Replace week one's opening balance with the real bank number.
- Move any invoice that didn't land. Don't delete it — shift it to next week and notice which customer it was.
- Add the week's new invoices and bookings, with the lag applied.
- Drop off the completed week, add week fourteen.
- Look at the closing-balance row and ask: did anything just move into the danger zone?
That's it. Fifteen minutes. The forecast's accuracy doesn't come from the sophistication of the spreadsheet — it comes from the fact that it's never more than a week out of date.
Gold nugget. Each Friday, before you update anything, glance at what you predicted this week's closing balance would be — then write the actual next to it. The gap between the two is your forecast error, and watching it shrink week by week is how you learn your own business's rhythms. Owners who track this one number for a quarter end up able to call their bank balance a month out within a few percent, and that confidence changes how they make every other decision: hiring, equipment, taking on the big job.
Where this fits
A cash flow forecast is one of the five reports every service business owner should be reading — the other four, and what each one is for, are covered in the five financial reports that actually matter. But if you currently read none of them, start here. The P&L tells you whether the business worked last month. The forecast tells you whether it survives the next three.
Build it this weekend. Thirteen columns, seven rows, one Friday habit. The first version will be wrong in a dozen small ways, and it will still be the most useful document in your business — because from now on, the bad week arrives with six weeks' warning, and six weeks is enough to do something about it.