Profitable businesses fail. They fail because profit is an accounting measure of a period and cash is what pays wages on Wednesday, and the two diverge for entirely ordinary reasons — customers paying late, stock bought ahead, tax due, loan principal repaid.
A cashflow forecast is the tool that closes the gap. Most small businesses either do not have one or have one that was built once for the bank and never updated.
Thirteen weeks, rolling
The format that works for most small businesses is a thirteen-week rolling forecast — a quarter ahead, updated weekly, with the horizon rolling forward so you always see the same distance.
Thirteen weeks is long enough to see a problem while you can still do something about it, and short enough that the numbers are real rather than aspirational. Annual budgets have their place and they do not tell you whether you can pay the GST bill in six weeks.
What goes in it
Cash in and cash out, by week, on the dates money actually moves — not the dates invoices are issued.
Receipts:
- Customer payments, timed by when each customer actually pays rather than by your terms. If a large customer reliably pays at 60 days on 30-day terms, model 60.
- Any other income — rent, interest, grants, tax refunds.
Payments:
- Wages and PAYE, on their actual dates.
- Supplier payments on their due dates.
- GST, provisional tax and other tax dates — these are known well in advance and are the most common cause of an unexpected shortfall.
- Loan repayments, including principal, which does not appear in the profit and loss.
- Rent, insurance, subscriptions and other fixed costs.
- Capital purchases.
- Drawings or dividends.
The two most commonly omitted items are loan principal and tax, which is precisely why businesses are surprised by both.
Make it honest, not optimistic
A forecast built on customers paying to terms and sales hitting target is a wish. The version that is useful models what actually happens:
- Use historic payment behaviour by customer, not contractual terms.
- Model committed and highly probable sales separately from hoped-for ones.
- Include the costs you know are coming but have not been invoiced yet.
Then run a downside: what if the largest customer pays a month late, or a major job is delayed by six weeks? If that scenario breaks you, you have found the thing to fix — and you have found it while there is time.
Using it weekly
The discipline is what makes it work. Once a week, update actuals against forecast, roll the horizon forward one week, and look at the low point.
Comparing actual against forecast also improves the forecast. Systematic errors show up quickly — a customer who always pays later than modelled, a cost category consistently understated.
Fifteen minutes a week is enough once it is built.
What to do when it shows a problem
The value is lead time. A shortfall visible eight weeks out has many solutions; the same shortfall discovered on the day has almost none.
Levers, roughly in order of how quickly they work:
- Collect faster. Chase overdue accounts, invoice immediately rather than monthly, ask for progress payments or deposits.
- Slow outflows. Negotiate with suppliers before the due date, which preserves the relationship in a way that missing payment does not.
- Defer discretionary spending — capital purchases, non-urgent maintenance.
- Reduce stock purchasing where inventory is high.
- Arrange funding — overdraft, invoice finance. Both are much easier to arrange with a forecast in hand and several weeks of notice.
- Talk to Inland Revenue early if tax is the issue. Instalment arrangements are available and are considerably easier to agree before the due date than after.
Why lenders ask for one
A business that arrives with a credible thirteen-week forecast, showing it understands its own cycle and has identified the pinch points, is assessed very differently from one that arrives with a request and no analysis.
The forecast is not just a lending document. It is evidence that someone is running the business rather than reacting to it.
Tools
A spreadsheet is entirely adequate and often better than software, because building it yourself forces you to understand the drivers. Most accounting packages have forecasting modules that pull actuals automatically, which reduces the updating effort once the structure is right.
business.govt.nz publishes free cashflow forecast templates and guidance for New Zealand businesses.
General information only, not financial advice.
