Cash flow forecasting: knowing how much there'll be, not how much there is
Today's balance decides nothing. What decides is how much there'll be on the 25th, when payroll falls due and the big supplier's payment is owed.
Having money and being able to spend it aren't the same thing
Looking at the bank balance answers a question of little use: how much there is now. The question that decides is a different one: how much will there be on the day I have to pay?
Profitable companies run out of cash all the time, and almost never for lack of margin: it's a mismatch of dates. You collect at 60 days and pay at 30, and in between there's a payroll.
Where the forecast comes from
The good news is that almost all the data already exists, scattered about:
- What you're going to collect: outstanding invoices with their due dates. Adjusted by how long that customer actually takes, not by what was agreed.
- What you're going to pay: outstanding supplier invoices, with their due dates.
- What's committed and not yet invoiced: approved purchase orders that will arrive.
- The recurring stuff: payroll, rent, utilities, taxes. The easiest to forecast and the heaviest.
If all of that lives in the same system, the forecast is a query. If it's in four places, it's a spreadsheet somebody rebuilds every week and that's out of date the same day.
The part that makes it useful: realism
A forecast that assumes everyone pays on the due date is useless. It's optimistic by design, and that's why nobody trusts it.
- Adjust for each customer's actual behaviour. If one always pays at 75 days having agreed 60, forecast them at 75.
- Separate the certain from the likely. An overdue invoice from a customer who isn't answering isn't a receipt: it's a hope.
- Look at scenarios. What happens if the big customer is two weeks late. That's what turns a forecast into a decision tool.
The value isn't in getting the number right: it's in seeing the gap three weeks ahead. With three weeks you negotiate a term or pull a payment forward. With three days, you take an expensive loan.