The 13-Week Cash Flow Forecast.

A 13-week cash flow forecast is a rolling, week-by-week projection of every dollar entering and leaving your bank account over the next quarter. It records money when it moves, not when you earn it. Its only job is to show you the week you run out of cash, before you get there.

Most businesses that fail are profitable on paper the month they fail. Cash flow forecasting is how you see the gap between the two.

Why 13 weeks

Thirteen weeks is one quarter, to the week. It is long enough to catch a quarterly tax payment, an inventory reorder, an annual insurance renewal and a customer who has quietly slipped to net 60. It is short enough that the numbers are still real.

Push it to 52 weeks and you are writing fiction. Pull it back to four and you cannot see far enough to change anything. A quarter is the horizon where a problem is still fixable: you can move a reorder, delay a distribution, chase a receivable, or call your bank while you still have leverage.

A cash flow forecast is not a P&L

This is the distinction that kills businesses. A P&L records revenue when you earn it. A forecast records money when it lands.

You can invoice $200,000 in March, book a superb month, and be unable to make payroll in April because nobody has paid you. The P&L is not lying. It is answering a different question. It tells you whether the business works. The forecast tells you whether it survives until it works.

Every line in a 13-week forecast is a bank movement. An invoice you sent is not cash. A sale on account is not cash. Amazon showing you a settlement figure is not cash until it clears.

What goes in it

Two blocks and a running balance. Open with your bank balance. Add what comes in. Subtract what goes out. Carry the closing balance into next week's opening. That is the entire mechanic.

Cash in: customer receipts, marketplace payouts net of fees, loan drawdowns, owner contributions.

Cash out: payroll, payroll taxes, inventory payments, inventory deposits, rent, advertising, software, debt service, estimated tax payments, insurance and annual renewals, owner distributions.

The lines everyone forgets

In seventeen years I have never seen a first-draft forecast that included all of these. They are the ones that turn a comfortable quarter into a crisis.

• Estimated tax payments. Quarterly, lumpy, and they land in a single week.
• Inventory deposits. Paid months before the goods arrive, and long before they sell.
• Annual renewals. Insurance, licences, software. Twelve months of comfort, then one bad week.
• Payroll tax remittances. Separate from payroll, on a different date.
• Merchant and marketplace holds. Amazon holds a reserve by default. That money is yours and it is not in your bank.
• Owner distributions. If you take them, model them. Most people do not.

The number that matters is not week 13

Here is a real shape, using round numbers. A business opens with $50,000, collects $40,000 a week, and spends $33,000 a week on payroll, rent and ads. Comfortable. Then in week 5 a $60,000 inventory deposit lands, and in week 7 an $18,000 estimated tax payment.

Week 13 closes at $63,000. Healthy. Look at the quarter as a whole and nothing is wrong.

But the minimum balance is $21,000, in week 7. That is the constraint. If payroll is $22,000, this business misses it, in a quarter that ends up. If a customer pays two weeks late, it misses it badly.

Watch the minimum, not the final week. Nobody has ever gone under in week 13.

How to roll it

Every Monday: replace week one with what actually happened, delete it, add a new week thirteen at the far end. It takes twenty minutes once the model exists.

That rolling step is the whole point, and the reason most forecasts are useless. A forecast built once is a document. A forecast rolled weekly is an instrument. The value is not in the projection. It is in watching the projection change, and knowing why.

Two rules keep it honest. Be pessimistic on inflows: if your terms are net 30, model net 45, because customers pay late. Be pessimistic on outflows: model the bill arriving early. And never smooth. A quarterly tax payment lands in one week, not spread across thirteen. Smoothing is how the dip disappears from view.

What to do when it dips

If the minimum is comfortable, do nothing. Roll it weekly and stop worrying about cash.

If it dips but stays positive, move one thing. Push a reorder ten days. Delay a distribution by a month. Chase the two invoices that matter.

If it goes negative, you have just found your problem eight weeks before it found you. That is enough runway to fix it. Call your bank in week two, when you have leverage and a plan, not in week twelve when you have neither.

Who maintains it

Someone has to. A model nobody updates is worse than no model, because it manufactures confidence.

This is the core of what fractional CFO services actually deliver, and it is the same work under the label outsourced CFO services. Not advice. An instrument, maintained weekly, that changes what you do. We build it, we roll it, and we tell you the week something breaks. You can see what that looks like in practice on our real estate holdings case study, where cash flow forecasting sat on top of a portfolio that was underwater and nobody could see why.

Questions, Answered.