Most growth-stage companies manage cash by checking the bank balance. That works until it does not, usually at the worst possible moment: a large receivable slips two weeks, a tax payment lands the same week as payroll, and a business that looked healthy on the income statement is suddenly scrambling for a bridge.
The 13-week cash flow forecast solves this. It is the most useful single financial tool for a company between $10M and $100M in revenue, and it is not complicated to build. What it requires is discipline: a weekly cadence and an honest set of inputs.
What a 13-week forecast actually is
A 13-week cash flow forecast is a rolling projection of every dollar expected to move in and out of the business, week by week, over the coming quarter. Each week you record what actually happened, drop the completed week, and add a new week 13 at the far end. The model never stops at a fixed end date; it rolls forward continuously.
The output that matters is the ending-cash balance for each of the 13 weeks. Read across that row and you see, in advance, the week your cash dips lowest and whether it crosses zero. That is information the bank balance alone will never give you, because the bank balance is a snapshot of today, not a projection of week 9.
Why build it direct, not from the P&L
There are two ways to forecast cash. The indirect method starts from projected net income and adjusts for non-cash items and working-capital changes. It is fast, but it inherits the accrual timing of the P&L, which is exactly the timing you are trying to see past.
The direct method builds from the actual expected events: this customer pays this invoice in week 4, this vendor gets paid in week 6, payroll runs every other Friday, the quarterly tax estimate is due in week 11. It takes more effort to assemble, but it reflects real cash timing. For a 13-week horizon, always build direct. The whole point is to capture the gaps between when revenue is earned and when cash arrives.
How to read it each week
Start with the ending-cash row. Find the low point and confirm it stays above your minimum operating threshold, including whatever buffer your board expects.
Then look at variance. Compare this week's actuals against what last week's forecast predicted for this same week. A single week off is noise. A pattern of collections coming in consistently below forecast is a signal: your days-sales-outstanding is drifting, and the model is catching it weeks before it would show up as a cash crunch.
That early-warning property is the real value. A company running a disciplined 13-week forecast rarely gets surprised by its own cash position, which means it can negotiate from strength rather than desperation.
Getting started
Build the first version in a spreadsheet. Pull your last quarter of actual receipts and disbursements, group them into recurring categories, and layer in the known one-time events. Once the structure is stable, automate the data pulls from your accounting system so the weekly refresh takes minutes, not hours.
The companies that do this well treat the weekly update as a fixed ritual, the same way they treat payroll. The cadence is what turns a forecast into foresight.