A 13-week cash flow forecast projects the cash a business will receive and pay, week by week, for the next quarter, starting from the cash it holds today. It is built from direct items such as customer receipts, supplier payments, payroll, tax and debt service, not from profit. It is rolled forward and checked against actual cash every week, so the lowest point in the quarter shows up early enough to act.
What is a 13-week cash flow forecast?
A 13-week cash flow forecast is a week-by-week projection of the cash a business will receive and pay over the next quarter, starting from the cash it holds today. It uses the direct method: actual receipts and payments, not profit adjusted for non-cash items. Its purpose is to show the lowest cash point in the coming weeks early enough to do something about it.
Thirteen weeks is one quarter. It is long enough to include several payroll cycles, tax payments and facility instalments, and short enough that most of the cash in it is already known: invoices issued, bills received, salaries agreed. Much beyond a quarter, a weekly forecast turns into guesswork and a monthly view works better.
Treasurers use it to plan funding and payment runs. CFOs use it to answer the board's question of how much cash the business will have, and when. A lender reviewing a facility may ask for one. It is worth running when cash is not tight too, because a forecast the team already trusts is the one it can act on quickly.
What goes into a 13-week cash forecast?
Each line needs a source and a timing rule. The table below is a practical starting set for a company in Saudi Arabia or the GCC; add or remove lines to match how your business pays and gets paid.
| Line | Where the figures come from | Which week it goes in |
|---|---|---|
| Opening cash | Bank balances on the start date, reconciled to the ledger | The opening balance of week 1 |
| Customer receipts | Open customer invoices, plus expected sales not yet invoiced | The week each customer is likely to pay, based on how it has paid before |
| Post-dated cheques | The cheque register | The cheque date, in place of the invoice the cheque settles |
| Supplier payments | Open bills, plus purchase orders due for delivery | Your payment run days, for the bills due by then |
| Payroll and social insurance | The payroll register and the headcount plan | The pay day, and the date contributions such as GOSI are due |
| Tax | VAT, withholding tax and zakat workings | The due dates in your filing calendar |
| Debt service | Schedules for loans, murabaha and tawarruq facilities | Each instalment date, principal and profit or interest |
| Rent, utilities and other recurring costs | Payments over the last few months | The usual payment day, or spread evenly when there is none |
| Capital spending and one-off items | Approved capex, dividends and settlements | The agreed date, labelled as a decision |
Keep new business that is not yet invoiced on its own line. It is the least certain number in the forecast, and anyone reading it should see how much of the closing cash depends on it.
How do you build a 13-week forecast, step by step?
Build it from the ledgers outward, one line at a time, and write down each assumption as you make it.
- Fix the calendarChoose the forecast date and the day each week ends, and keep both fixed. Week 1 begins the day after the forecast date. Use the same weeks for every company in a group.
- Start from reconciled cashTake the balance of every bank account on the forecast date and agree it to the ledger. A forecast that starts from an unreconciled number is wrong in every week. The bank reconciliation guide covers this step.
- Lay out customer receiptsExport open receivables by invoice. Place each invoice in the week the customer is likely to pay, not the week it falls due. Keep invoices that are already overdue on a separate line until someone confirms a date.
- Lay out supplier paymentsExport open payables and put each bill in the payment run that will settle it. Bills already past due belong in week 1, unless you have agreed a later date with the supplier.
- Add payroll, tax and debtPlace salaries on the pay day, social insurance on its due date, tax payments on the dates in your filing calendar and each facility instalment on its schedule date.
- Add recurring costs and decisionsEstimate recurring costs from recent months. Add capital spending, dividends and other one-off items only once someone has approved them, and label them.
- Calculate and testEach week's closing cash is its opening cash plus receipts minus payments. Compare every week with your minimum cash buffer and with the undrawn headroom on your facilities.
- Give each line an ownerReceivables, payables, payroll and treasury each own their lines and confirm them every week. The forecast is a shared document, not one person's file.
How should you time customer receipts?
Time receipts by how each customer actually pays, not by the due date on the invoice. Receipts are usually the largest and least certain line, and for most customers the due date is optimistic.
The simplest method is to measure, for each customer, the average number of days between the due date and the day payment arrived on its recently settled invoices, then move each open invoice by that many days. Use the due date only when a customer has too little history to measure.
Some receipts need their own judgement: customers paying against milestone certificates, customers in dispute, and large balances that are already overdue. Show overdue receivables below the forecast with their total, rather than assuming they will arrive next week.
What does a 13-week forecast look like?
The forecast is a table with one column per week, receipts and payments by line, and a running closing balance. The extract below shows the first four weeks of a simplified forecast, with totals in place of the lines.
| Week | Opening cash | Receipts | Payments | Closing cash |
|---|---|---|---|---|
| Week 1 | 4,000,000 | 1,200,000 | 1,500,000 | 3,700,000 |
| Week 2 | 3,700,000 | 900,000 | 2,600,000 | 2,000,000 |
| Week 3 | 2,000,000 | 1,400,000 | 800,000 | 2,600,000 |
| Week 4 | 2,600,000 | 1,100,000 | 1,900,000 | 1,800,000 |
In this example, payroll and a VAT payment fall in week 2 and a facility instalment in week 4. If the company keeps a minimum buffer of SAR 2,000,000, week 4 falls below it. Seen four weeks ahead, the options are still open: chase the two largest overdue customers, move a discretionary payment to week 5, or arrange a short drawing on the facility.
How do you run and review the forecast every week?
Roll it forward on the same day every week, and compare last week's forecast with what actually happened before you change anything.
- Drop the week that has ended and add a new week 13.
- Compare the forecast closing cash for the week just ended with the actual bank balance, and explain the difference line by line.
- Sort each difference into timing (the cash moved to another week) or permanent (the cash will not come, or was not needed). Timing differences reverse; permanent ones change the outlook.
- Update the assumptions behind the largest differences, such as a customer's payment pattern or a supplier's terms.
- Save each week's version, so you can measure accuracy over time by line and by how many weeks ahead.
- Agree actions in a short weekly meeting: who chases which customer, which payments wait, and whether to draw on a facility or move cash between group companies.
Accuracy improves when someone owns it. Track how far each week's forecast closing cash was from the actual balance, and look at the few lines that cause most of the misses.
What mistakes make a cash forecast unreliable?
Most unreliable forecasts fail in the same few ways.
- Starting from a bank balance that has not been reconciled to the ledger.
- Placing receipts on due dates, which assumes every customer pays on time.
- Forgetting items that come once a quarter or once a year: zakat, licence renewals, insurance premiums, bonuses and end-of-service payments for leavers.
- Mixing profit and cash. Depreciation, accruals and provisions are not cash; the payments behind them are.
- Counting cash twice in a group. Money one company pays another does not change the group's cash, so take it back out of the group total. The guide to group consolidation and intercompany eliminations explains why.
- Overwriting the file each week, so there is no record to measure accuracy against.
- Keeping the forecast in one person's spreadsheet with links nobody else can follow. The comparison of Naarvin and spreadsheets covers when a spreadsheet is still the right tool.
Questions people ask
How is a 13-week cash forecast different from a budget?
A budget plans revenue, costs and profit for the year, usually by month, and is approved once. A 13-week forecast tracks cash only, week by week, and changes every week as invoices are issued and paid. You need both: the budget sets the plan, and the cash forecast shows whether you can fund it this quarter.
Should we use the direct or the indirect method?
Use the direct method for a 13-week forecast: receipts and payments by line. The indirect method, which starts from profit and adjusts for working capital, suits longer monthly or annual forecasts, where individual invoices are not yet known.
How accurate should the forecast be?
There is no standard target. Measure the difference between forecast and actual closing cash for each week and watch the trend. Near weeks should be close, because most of their cash is already known; later weeks will miss by more, and measuring tells you by how much.
How often should we update it?
Every week, on the same day, once the bank balances are in. When cash is very tight, add a daily cash position for the next two weeks alongside the 13-week view.
How should a group of companies forecast cash?
Build the forecast for each company in its own currency, then combine them in the group's currency. Take out payments between group companies so the group's cash is not overstated, and keep each company visible, because cash in one company is not always free to move to another.
Can we run a 13-week forecast in a spreadsheet?
Yes, and many teams start there. It gets hard to maintain with many companies, currencies or customers, because each week means new exports and rebuilt links. At that point a forecast that reads the ledgers directly saves the weekly rebuild.