You should know you can make payroll in week nine. Not find out in week nine. The 13-week cash flow forecast is the one tool that gives you that lead time, and most growing companies still do not run it. Here is exactly how it works, and why thirteen weeks is the number.
You can be profitable on the P and L and still get caught short on cash. Profit is an accounting idea. Cash is a calendar problem. A customer who owes you $80,000 is profit today and nothing in the bank until they pay. The 13-week forecast lives on the calendar, so it answers the only question that keeps founders up at night: do we have enough in the account on the day the money is due?
Why thirteen weeks
Thirteen weeks is one quarter, and the window is deliberate. It sits between two reports that cannot do this job. An annual budget is a strategy document, built in months and almost always wrong by week three, and it stays silent on whether you can cover this Friday. Your bank balance is the opposite problem, perfectly accurate and useless for deciding anything, because by the time the number scares you the moment to act has gone.
Thirteen weeks is near enough to forecast accurately, since you know who owes you, who you owe, and when payroll and taxes hit. It is far enough out to still act: collect a late invoice, delay a purchase, draw on a line, or hold a hire.
What goes in it
The model is simple on purpose. Four rows, week by week.
Starting cash
The actual balance in your operating accounts at the start of week one. Not the figure in your accounting software, the figure in the bank, reconciled. Every week after that, starting cash is just the prior week's ending cash. This is the only number you do not forecast.
Cash in
Money landing in the account, timed to when it actually arrives. This is the part most spreadsheets get wrong. You forecast collections, not bookings: a signed deal is not cash, an invoice is not cash, cash is the day the payment clears. So you walk your receivables and place each payment in the week you realistically expect it, based on the customer's habits, not their terms. The client who is "net 30" but always pays in 50 goes in week seven, not week four.
Cash out
Everything leaving the account, in the week it leaves. Payroll and the taxes that ride with it. Accounts payable, vendor by vendor. Rent, software, and recurring bills. Estimated income and sales taxes on their real due dates. Debt service. And owner draws, which founders love to leave off and then wonder where the cash went.
Ending cash, per week
Starting cash, plus cash in, minus cash out. That single number, repeated down all thirteen columns, is the entire point: your bank balance on the last day of each of the next thirteen weeks, before it happens. The week it turns red, or thin, is the week you have time to fix.
How to build and run it
A forecast you build once and never touch is just a budget, and a worse one. The power is in the rhythm:
- Build the grid. Thirteen columns, one per week. Rows for starting cash, cash in, cash out, and ending cash. A clean spreadsheet is plenty to start.
- Anchor it in actuals. Fill week one from your real bank balance and your named receivables and payables. Specifics beat averages.
- Forecast on real timing. Place every dollar in the week it truly moves, using each customer's payment behavior and each bill's real due date, not the paper terms.
- Re-forecast every week. Drop the week that just closed, add a fresh week thirteen, and true up every column against what actually happened. It always stays thirteen weeks long. That is what "rolling" means.
- Make one decision per update. The forecast is not a report to admire. Every update should produce one action: chase that invoice, push that purchase, or do nothing on purpose.
Tied to actuals and re-run weekly, the forecast gets sharper, not staler, because you see week over week who actually paid when.
A budget tells you where you are going. A 13-week forecast tells you whether you arrive alive.
What it tells you that your P and L cannot
Your income statement can show a great month and hide a cash crunch two weeks out. The forecast tells you the precise week things get tight. Not "cash is a bit low," but "week eight dips to $42,000 and a $60,000 tax payment is due that Friday." A number you can act on, not a vague worry.
It also tells you the cash effect of a decision before you make it. Put a $9,000-a-month hire into the model, slide one large receivable from week five to week nine, and see whether you still clear payroll in week six. And it gives you runway in plain weeks, not a fuzzy "a few months."
Want your runway as a hard number, in 30 seconds? Drop in your cash on hand and monthly burn. See the exact week the line hits zero, then count back to the week you have to act.
Open the Runway Calculator →Key takeaways
- Thirteen weeks is one quarter: close enough to forecast accurately, far enough to act.
- Forecast by the week cash actually moves, not by booking dates or paper terms.
- It is rolling. Re-forecast weekly, true up to actuals, and make one decision each time.
- It answers what the P and L cannot: the exact week things get tight, and your runway in plain weeks.
You can build this yourself, and you should at least understand it. But running it every week, truing it to actuals, and turning it into one clear decision is a job, and it is the seat a Rainier CFO fills. You get the rolling forecast, a clean monthly close in plain English, a dashboard of your key numbers, and a standing meeting where someone senior tells you what to do about them. See what a Rainier CFO does or how the engagement is priced.