Almost every business has tried to build a cash flow forecast at some point — usually the week a tight month caught someone by surprise. A spreadsheet gets built, the next few months get modelled, everyone feels better, and then it quietly stops being updated once the immediate scare passes. Six months later it's exactly as useless as not having one at all, because the inputs are stale and nobody trusts the output enough to act on it. A cash flow forecast that actually earns its keep isn't a one-off exercise done under pressure. It's a living model, rebuilt on a rhythm, that's accurate enough to change a real decision — when to collect harder, when to hold off on a hire, when to go and talk to the bank before the problem is already in the account.
The mechanics are simpler than most owners expect, which is partly why the discipline around maintaining it matters more than the model itself. Start with the actual cash balance today, not a budgeted one. Lay out every inflow — customer receipts, by expected payment date rather than invoice date — and every outflow, in the same terms. The near weeks should be modelled weekly, because that's where the real risk sits; the months further out can be monthly, since precision that far ahead is mostly false confidence anyway. A rolling 13-week view, updated weekly or fortnightly rather than rebuilt from scratch each time, is the format that tends to actually get used, because it never goes stale enough to feel untrustworthy.
The inputs that make or break the forecast
Revenue timing is where most forecasts quietly fail, usually through optimism rather than dishonesty — modelling when a customer is expected to pay rather than when they actually tend to pay. The fix is to use the business's own real debtor history, not the payment terms printed on the invoice. On the outflow side, the Australian-specific obligations are the ones that catch people out because they land in lumps rather than smoothly: BAS and GST, PAYG instalments, superannuation guarantee payments, payroll tax where it applies, and loan repayments, alongside the more obvious rent, payroll and supplier terms. A forecast that only models the smooth, predictable costs and leaves the lumpy ones to surprise the account later isn't really a forecast — it's a budget wearing a forecast's name. This is the same gap that explains why a genuinely profitable business can still run short of cash: the P&L smooths timing away entirely, and a forecast is only useful to the extent it puts the timing back in.
Why the habit matters more than the spreadsheet
The single biggest reason a forecast stops being useful isn't a modelling error — it's that nobody compares what actually happened against what was forecast, so the gaps never get corrected and the next version is built on the same bad assumptions as the last one. Reviewing actual cash movements against the forecast every month, and asking specifically why the variance happened, is what turns a rough first attempt into a model the business can actually rely on six months later. It's the same discipline behind the handful of forward-looking KPIs worth tracking properly — debtor days and cash runway in particular feed directly into how reliable the forecast's near-term weeks turn out to be.
A cash flow forecast isn't a compliance document and it isn't really about predicting the future with precision — no SME forecast survives contact with a surprise invoice or a slow-paying client fully intact. Its real job is narrower and more useful than that: giving enough warning, far enough ahead, that a tight month becomes a planned-for month instead of a crisis discovered the week it arrives.