Cash Flow Forecasting & Decision Modeling

Why a 13-week cash flow forecast

Most owners of $1M–$5M businesses run finance off the bank balance. It works until it doesn't — a good growth month front-loads costs, a big customer pays late, a tax bill and a payroll run collide, and suddenly the balance that looked healthy on Monday is uncomfortably thin on Friday. The 13-week forecast exists to make that visible before it happens.

Thirteen weeks is one quarter. It's long enough to capture the events that actually move cash — several payroll runs, a quarterly tax payment, the real payment behavior of your slowest customers — and short enough that you can forecast weekly timing with confidence. It is a cash model, not an accrual one: the only question each cell answers is what will hit the bank, and in which week?

Definition

Rolling 13-week cash flow forecast

A weekly projection of beginning cash, expected receipts, and all outflows across the next 13 weeks, updated every week by dropping the week that ended and adding a new thirteenth week. It predicts your actual bank position — not booked revenue — so you can see shortfalls a quarter in advance.

How to build one, step by step

You can build the first version in an afternoon with a spreadsheet: thirteen columns across the top, one per week, and the rows below in order. Work top to bottom.

1. Start with today's real cash balance. Your true available cash across all operating accounts, net of anything already committed but uncleared. This is week zero's beginning cash and the anchor for the whole model.
2. Lay out expected receipts by timing. For each week, enter the cash you expect to actually collect — driven by your AR aging and how each customer really pays. A 45-day payer lands in the week they pay, not the week you invoiced.
3. Enter payroll and payroll taxes on their real dates. Map every payroll run and its tax deposits into the exact weeks they clear. Payroll is usually the largest, least flexible outflow, so its timing is what makes the model trustworthy.
4. Schedule accounts payable and operating costs. Add vendor payments, rent, software, and recurring costs in the weeks you intend to pay them, working from AP aging and terms rather than spreading evenly.
5. Add taxes, debt service, and owner draws. Layer in estimated tax payments, loan and interest payments, leases, and distributions on their due dates. These lumpy, forgettable items cause most surprise shortfalls.
6. Calculate ending cash and carry it forward. Beginning cash plus receipts minus outflows equals ending cash; that figure becomes next week's beginning cash. Now you have a running line of your position every week for a quarter.
7. Update weekly and read the low point. Each week, drop the oldest week, add week 13, and replace estimates with actuals. Watch the lowest ending-cash week — that trough is the number that should drive your decisions.

What a good cash model includes

• Beginning cash that ties to real bank balances, not book balances.
• Receipts by expected payment date, built from AR aging and actual customer behavior.
• Every material outflow — payroll, taxes, AP, rent, debt service, leases, and owner draws — on its real date.
• A visible weekly ending-cash line and a flagged low point across the quarter.
• Actuals replacing estimates each week, so accuracy improves rather than decays.
• A minimum cash threshold — the balance you refuse to go below — drawn as a line you can see.

From forecast to decision modeling

A forecast tells you where cash is heading on the current path. Decision modeling asks a sharper question: what happens to that path if we make a specific move? This is where the model earns its keep. Before you commit real money — a hire, a second location, a price change, a piece of equipment — you run the decision through three versions of the future.

• Base case. Your honest, most-likely assumptions. Not optimistic, not defensive — the number you'd actually bet on.
• Best case. Things break your way: the pipeline lands faster, the new hire ramps quickly, collections hold.
• Worst case. The one that matters most. Revenue comes in slow, the ramp takes longer, a customer stretches payment. Can you survive it without breaching your minimum cash line?

Take a $90k/year hire. In the model that's roughly $8k–$9k per month all-in landing in specific weeks, while the revenue they enable shows up later and less certainly. Model the hire and you can see exactly how many months of runway it costs before it pays back — and whether the worst case still clears your minimum. That's the difference between deciding on nerve and deciding on numbers. It's the same discipline behind preparing for growth, capital, or exit, and one of the clearest signs a business is ready for strategic finance.

None of this requires expensive software or a full-time hire. It requires the model, the weekly rhythm, and the judgment to read it — which is exactly the work a fractional CFO owns. If you're weighing whether that's your next step, start with when to hire a fractional CFO or the readiness assessment.