Clients rarely run out of profit. They run out of cash, usually in a specific week, when payroll, rent and a tax remittance all land before a big customer pays. A monthly P&L won't show that week coming. A 13-week cash flow forecast will.
The forecast is a weekly table of cash in and cash out for the next quarter, built from the bank balance, the receivables and payables agings, and a short list of recurring payments. Updated every week, it's one of the most useful things an accounting firm can give a client. This is the method we built 13-Week Cash Forecast around. It works just as well in a spreadsheet, and our free 13-week cash flow forecast template is a ready-made starting point.
How the forecast is laid out
Thirteen columns, one per week, and the same rows every time:
| Row | What goes in it | Where it comes from |
|---|---|---|
| Opening cash | Cash at the start of the week | Week 1: bank balances. After that: last week's ending cash |
| Receipts | Customer collections, new invoicing, other cash in | AR aging, sales run-rate, known one-offs |
| Disbursements | Suppliers, payroll, rent, loans, taxes, credit card | AP aging and recurring items |
| Net cash flow | Receipts minus disbursements | Calculated |
| Ending cash | Opening cash plus net cash flow | Calculated, and becomes next week's opening cash |
| Minimum cash | The floor the client has agreed to | Agreed with the client |
It's a cash forecast, not a P&L. Depreciation and accruals don't appear. What matters is the week money actually leaves or reaches the bank.
Building it, step by step
1. Start from the bank
Week 1's opening cash is the sum of the client's bank balances, all as of the same date. Balances from different days double count or miss transfers between accounts. If the client has foreign-currency accounts, convert them at that day's rate and note it in the assumptions.
2. Place collections from the AR aging
Export the Aged Receivables Detail from QuickBooks Online or Xero. For each open invoice, the expected week is the due date plus that customer's average days late, worked out from their payment history. Customers who reliably pay 20 days late will keep doing it, and the forecast should say so.
Leave out invoices that are 90 or more days overdue unless there's a specific reason to expect them, such as a payment plan. They're far more likely to be written off than collected this quarter.
3. Add sales that haven't been invoiced yet
The aging covers only invoices already sent. For the rest of the quarter, add a weekly new invoicing amount based on the client's run-rate, collected after their average days to pay. For a client invoicing $16,000 a week whose customers pay in about 35 days, the first of those collections lands five weeks out.
4. Place supplier payments from the AP aging
Export the Aged Payables Detail and place each bill on its due date or, if the client pays suppliers in a weekly run, on that run's day. Ask the client which bills they plan to stretch. That's a decision for them, but the forecast should reflect it.
5. Set up recurring items once
Most of a forecast's accuracy comes from items that aren't in either aging. Set each one up with an amount and a schedule, and let it repeat:
- Payroll and source deductions, usually bi-weekly. Watch for months with three pay dates.
- Rent and loan payments, on their monthly date.
- Tax remittances: sales tax, payroll remittances and income tax instalments.
- The credit card payment, on the statement due date. Use the statement payment, not the individual charges, or you'll double count anything that's also in the AP aging.
- Subscriptions and other fixed costs, as a weekly or monthly amount.
6. Add the one-offs
Equipment purchases, annual insurance premiums, bonuses, owner draws, a loan advance coming in. Ask the client every week: "Is anything big coming in or going out that I don't know about?"
7. Set the minimum and find the lowest point
Agree a minimum cash balance with the client. A practical starting point is about two weeks of payroll and fixed costs, adjusted for lender covenants and how lumpy the business is. Then find the lowest ending cash over the 13 weeks, the week it happens, and the first week, if any, that falls below the minimum. That's the headline of every update.
8. Write down the assumptions
Average days to pay, the new invoicing run-rate, which invoices were left out, the timing of any purchase. A short list beside the forecast lets the client challenge the right number, and lets anyone on your team update it the same way next week.
A worked example
Here's the first part of a forecast for a fictional client, Fieldstone Joinery Ltd., prepared on Monday, November 2. Opening cash across its two bank accounts is $84,500, and the agreed minimum is $40,000.
The largest open invoices in the AR aging are placed like this:
| Customer | Open | Due | Avg days late | Expected |
|---|---|---|---|---|
| Alder Homes | $18,400 | Oct 28 | 14 | Nov 11 (week of Nov 9) |
| Brookline Developments | $9,600 | Nov 4 | 7 | Nov 11 (week of Nov 9) |
| Corbett Kitchens | $12,200 | Nov 10 | 21 | Dec 1 (week of Nov 30) |
| Dunmore Interiors | $4,300 | Jun 30 | n/a | Left out: over 90 days |
With the smaller invoices, the AP aging and the recurring items added, the first six weeks look like this:
| Week of | Nov 2 | Nov 9 | Nov 16 | Nov 23 | Nov 30 | Dec 7 |
|---|---|---|---|---|---|---|
| Opening cash | $84,500 | $66,800 | $84,800 | $70,500 | $68,300 | $50,200 |
| Collections (AR aging) | $6,800 | $28,000 | $7,500 | $3,900 | $12,200 | $2,100 |
| New invoicing | $0 | $0 | $0 | $0 | $0 | $16,000 |
| Payroll | −$14,500 | $0 | −$14,500 | $0 | −$14,500 | $0 |
| Suppliers (AP aging) | −$5,200 | −$7,800 | −$3,400 | −$6,100 | −$4,700 | −$5,000 |
| Rent | −$4,800 | $0 | $0 | $0 | −$4,800 | $0 |
| Loan payment | $0 | −$2,200 | $0 | $0 | $0 | $0 |
| Credit card | $0 | $0 | −$3,900 | $0 | $0 | $0 |
| Sales tax remittance | $0 | $0 | $0 | $0 | −$6,300 | $0 |
| Ending cash | $66,800 | $84,800 | $70,500 | $68,300 | $50,200 | $63,300 |
The week to watch is November 30. Payroll, rent and the sales tax remittance all go out that week, and ending cash drops to $50,200, just over $10,000 above the minimum. Collections are thin in the weeks of November 16 and 23 because the October invoices have mostly been placed already, and new invoicing doesn't start arriving until December 7, five weeks after it starts.
Nothing breaches the minimum, but it's close enough to tell the client: "Cash is lowest in the week of November 30, about $10,000 above your minimum. If Corbett Kitchens pays late, that week gets tight."
Rolling it forward
The forecast is only useful if it's updated every week, on the same day. On Monday, November 9, the week of November 2 is over:
- The new bank balances become the actual ending cash for the week of November 2: $63,000, against $66,800 forecast.
- Each category is compared with what was forecast, and the $3,800 difference is explained.
- The week of November 2 drops off, and a new week 13, the week of February 1, is added.
- Collections are re-placed from the new AR aging, supplier payments from the new AP aging, and recurring items carry on as set.
The comparison is the part that builds trust, and the part most firms skip. We cover it in Forecast vs actual: running the weekly cash forecast update.
Common mistakes
- Placing collections on the due date. Almost nobody pays exactly on time. Use each customer's history.
- Counting card charges and the card payment. Use the statement payment only.
- Forgetting non-aging items. Payroll remittances, tax instalments and loan payments are the usual misses.
- Keeping old AR in the forecast. A 120-day-old invoice in next week's collections makes every week look better than it is.
- Building it once. A 13-week forecast that isn't rolled forward is out of date within a fortnight.
Frequently asked questions
What is a 13-week cash flow forecast?
A 13-week cash flow forecast projects a business's cash receipts and payments week by week for the next quarter. Each week starts with the previous week's ending cash, so it shows when cash will be tight, not just whether the quarter is profitable. It's updated weekly, adding a new week as each one passes.
Why 13 weeks?
Thirteen weeks is one quarter. It's long enough to see payroll cycles, rent, loan payments and quarterly tax instalments, and short enough to forecast from real invoices and bills rather than estimates. Monthly forecasts hide the timing problems that a weekly view catches.
How do I forecast customer collections?
Start from the AR aging. Place each open invoice in the week it's likely to be paid: its due date plus that customer's average days late from their payment history. Leave out very old invoices (90 days or more overdue) unless there's a reason to expect them, and add sales not yet invoiced as a weekly amount collected after the average days to pay.
How do I set a minimum cash balance?
Agree it with the client. A practical starting point is about two weeks of payroll and fixed costs, adjusted for how lumpy the business is and any lender covenants. The forecast then flags the first week ending cash is expected to fall below it.