Why profitable companies run out of cash
Profit and cash are not the same thing. You can book a large sale, recognize the revenue, and still be unable to make payroll because the customer pays in 45 days. The businesses that fail rarely fail from lack of profit — they fail from lack of cash at the wrong moment.
The 13-week rolling forecast
Thirteen weeks (one quarter) is the sweet spot: long enough to see trouble coming, short enough to forecast with real accuracy. Each week:
- Start with your actual bank balance.
- Add expected inflows — receivables due, based on each customer's real payment behavior, not the invoice terms.
- Subtract expected outflows — payroll, rent, AP due, taxes, loan payments.
- Carry the ending balance forward as next week's opening balance.
Forecast from behavior, not terms
A customer on Net 30 who always pays in 47 days should be modeled at 47 days. Use your own payment history per customer. This single adjustment is the difference between a forecast that's directionally useful and one that's dangerously optimistic.
Watch the danger weeks
The forecast's job is to surface the week where the balance dips below your minimum cash buffer — before it happens. When you see it three weeks out, you have options: accelerate collections, delay a discretionary purchase, or draw on a line of credit. When you see it the day of, you have none.
Build a cash buffer
Target a minimum operating buffer of 4–8 weeks of fixed costs. Treat that floor as untouchable in the forecast. The buffer turns a cash crisis into a cash inconvenience.
Revisit weekly
A forecast is a living document. Update actuals every Monday, roll the window forward one week, and compare last week's forecast to what actually happened. The gap between forecast and actual is where you learn how your business really behaves.