A 13-week cash flow forecast is a weekly view of money in and money out for the next quarter, built from what you know is coming rather than from your budget. Thirteen weeks is the sweet spot — long enough to see a shortfall while you can still do something about it, short enough that the numbers are real rather than guesses.
It is the single most useful financial tool a small business can build, and it takes about two hours to set up the first time.
Why 13 weeks
Shorter than 13 weeks and you cannot act. Finding out on Monday that Friday is a problem leaves you with expensive options only.
Longer than 13 weeks and you are forecasting rather than planning. Week 30 is a guess, and guesses reduce trust in the whole document.
Thirteen weeks is one quarter. Long enough to cover a VAT cycle, three payroll runs, a provisional tax payment and most supplier terms. Short enough that most of the numbers are things you actually know.
The structure
Columns are weeks. Rows are the lines below. Build it in a spreadsheet — this is one job spreadsheets do better than accounting software.
Opening cash
The actual bank balance at the start of the week, across all accounts. Use the real number, not the ledger balance.
Money in
| Row | How to populate it |
|---|---|
| Receipts from existing invoices | Your debtors list, scheduled by when each customer actually pays, not by due date |
| Expected new sales | Only what you are confident of. Be conservative |
| Other income | Interest, rebates, insurance claims |
| Funding | Loan draws, shareholder contributions |
| VAT refunds | If you are in a refund position |
The debtors point is the whole discipline. Do not schedule an invoice due on the 15th to arrive on the 15th if that customer has never once paid on time. Schedule it when they actually pay. Your forecast is only as good as this row.
Money out
| Row | How to populate it |
|---|---|
| Payroll | Fixed dates, known amounts |
| PAYE and UIF | 7th of each month |
| VAT payment | Last business day of the month following your tax period |
| Provisional tax | 31 August and 28 February for a February year-end |
| Rent | Fixed |
| Suppliers | From your creditors list, scheduled by due date |
| Loan and finance repayments | Fixed |
| Insurance, subscriptions, utilities | Fixed and recurring |
| Owner drawings | Be honest about these |
| Capital purchases | Anything planned |
Closing cash
Opening + in − out. This becomes next week's opening.
The row that matters most
Add a "lowest point in the next 13 weeks" cell at the top of the sheet. That single number is what the whole document exists to produce. If it is negative in week 7, you have six weeks to fix it — which is a completely different situation from discovering it in week 7.
Populating it the first time
1. Get the actual bank balance today, across every account.
2. Export your debtors age analysis. Schedule each invoice into the week that customer realistically pays. Look at their last six payments, not their terms.
3. Export your creditors list. Schedule each by due date.
4. Add every fixed and recurring payment. Payroll, rent, loans, insurance, subscriptions. These are the easy rows and they are usually the largest.
5. Add the tax dates. PAYE on the 7th, VAT at month-end, provisional tax in August and February. These are the payments that cause most surprises, and they are entirely predictable.
6. Add new sales conservatively. If you are unsure, leave them out. A forecast that proves pessimistic is far less damaging than one that proves optimistic.
7. Look at the lowest point.
The weekly discipline
This is what separates a forecast that works from a spreadsheet nobody opens.
Every week, on the same day, spend twenty minutes:
Replace last week's forecast with actuals. What actually came in, what actually went out.
Explain the variance. Where the forecast was wrong, understand why. This is how the forecast gets more accurate.
Roll forward a week. Drop the week just completed, add a new week 13.
Update the debtors and creditors rows with anything that has changed.
Look at the lowest point again.
The variance review is where the value is. After six weeks you will know which customers pay late, which costs you underestimate, and how reliable your sales pipeline actually is.
What to do when the lowest point goes negative
Having six weeks of warning gives you options that having six days does not.
Accelerate money in:
Call the three largest overdue debtors personally
Offer a small settlement discount for immediate payment on aged invoices
Invoice work in progress now rather than at completion
Take deposits on new work
Delay money out:
Talk to suppliers before you miss a payment, not after. A supplier given notice is usually accommodating; one who discovers a missed payment is not
Defer non-essential capital purchases
Reduce or defer owner drawings
Deal with SARS deliberately. If a VAT or PAYE payment is the problem, file the return on time regardless and apply for a payment arrangement. Late filing and late payment are separate failures with separate penalties, and an arrangement is far easier to obtain before enforcement than after. See how SARS penalties and interest are calculated.
Arrange facilities early. A bank will lend to a business that shows a forecast and a plan. It is considerably less willing when the account is already overdrawn.
The two mistakes that ruin it
1. Scheduling debtors by due date rather than payment behaviour. This makes the forecast optimistic in exactly the way that matters, and it is the reason most first attempts fail.
2. Forgetting the tax money is not yours. Your bank balance includes VAT and PAYE owed to SARS. If your forecast treats it as available cash, the forecast is wrong by exactly that amount. See why your bank balance is not your profit.
Frequently asked questions
How do I forecast cash flow for a small business? Build a weekly spreadsheet covering 13 weeks, with opening cash, receipts scheduled by when customers actually pay, payments scheduled by due date including payroll and tax dates, and closing cash rolling into the next week. Update it weekly with actuals and roll it forward.
Why 13 weeks and not 12 months? Thirteen weeks is long enough to see a shortfall while you can still act on it, and short enough that most of the numbers are things you actually know rather than guesses. It also covers a full VAT cycle, three payroll runs and a provisional tax payment.
What should be in a cash flow forecast? Opening cash, receipts from existing invoices and expected new sales, other income and funding, then payroll, PAYE and UIF, VAT, provisional tax, rent, suppliers, loan repayments, recurring costs, owner drawings and capital purchases, ending with closing cash.
How often should I update a cash flow forecast? Weekly, on the same day. Replace the completed week with actuals, understand the variance, roll forward a new week 13, and check the lowest point. Twenty minutes a week is what makes it accurate.
What do I do if my forecast shows I will run out of cash? Act on the warning while you have weeks rather than days. Accelerate collections from your largest overdue debtors, take deposits, defer non-essential purchases and drawings, talk to suppliers before missing a payment, and if a SARS payment is the issue, file the return on time and apply for a payment arrangement.
Is a cash flow forecast the same as a cash flow statement? No. A cash flow statement is a historical report forming part of annual financial statements, reconciling profit to cash for a past period. A cash flow forecast looks forward, weekly, and exists to help you make decisions.
Set it up once, use it every week
The forecast is straightforward. The discipline is the difficult part, and it is what turns a spreadsheet into an early warning system.
Smartbook builds the first 13-week forecast with clients from their actual debtors, creditors and tax obligations, then keeps it current alongside monthly management accounts — so the lowest point in the next quarter is always a number you know.
Last reviewed: 26 July 2026. Written by the Smartbook team — SAIPA and SAICA accredited, SARS registered tax practitioners. General guidance, not advice on your circumstances.