Cash flow forecasting

Profitable companies go bankrupt. Unprofitable ones can run for years. The difference is cash, and whether you saw the crunch coming. Cash flow forecasting is the most underrated operator skill. It takes an afternoon to set up and saves businesses every month.

Cash is not profit

A reminder of why this matters:

• You invoice a customer $100K. The P&L shows $100K revenue. Bank shows $0. Net 60 terms mean cash arrives in 2 months.
• You sign an annual contract for $120K collected upfront. Cash arrives day one. P&L recognizes $10K/month.
• You buy a server for $50K. Cash out day one. P&L shows $50K ÷ 3 years = $16K/year in depreciation.

All three distort the relationship between what's "earned" and what's in the bank. A cash flow forecast is how you translate between them.

The 13-week cash flow

The standard operator tool. A rolling weekly forecast of cash in, cash out, and ending balance for the next 13 weeks (one quarter). Updated every Monday.

Week:                W1    W2    W3    W4    ...   W13
Starting cash:       500   610   640   590
+ Receivables:       250   300   200   180   ...
+ New sales cash:    200   150   250   300
- Payroll:           (180) (0)   (180) (0)
- Rent:              (50)  (0)   (0)   (0)
- Vendors:           (80)  (120) (90)  (110)
- Taxes / other:     (30)  (0)   (230) (0)
Ending cash:         610   640   590   960   ...

Inputs, receivables

Pull your accounts receivable aging. For each invoice, assign a collection week based on terms + history:

• Net 30 invoice sent Week 1 → cash Week 5 or 6 (customers pay late)
• Net 60 invoice sent Week 1 → cash Week 9 or 10
• Habitual slow-payer → add 2 weeks

Inputs, payables

The out-flows are more predictable:

• Payroll, biweekly on fixed dates. Known to the penny.
• Rent, monthly on the 1st
• Vendors, scheduled by your AP terms
• Taxes, quarterly estimates, payroll taxes, sales tax remittance
• Known one-offs, annual software renewal, a piece of equipment

The cash runway calculation

If the business is burning cash, runway = current cash ÷ monthly burn. The forecast turns this from a rough estimate into a specific date.

Example. Current cash $800K. Monthly burn $100K. Rough runway = 8 months. But the forecast shows a $200K tax payment in month 3 and a $100K annual software renewal in month 5. Actual runway = 6 months, not 8. Six-month plans look very different from eight-month plans.

Stress testing

Once the base forecast is set up, run three cases every time you update:

1. Base, what you expect
2. Downside, top customer delays payment 60 days, no new deals close this quarter, one bad-debt write-off
3. Disaster, top customer churns, new sales drop 50%, one key payable gets called immediately

The question isn't "will we hit base?", it's "where are we on disaster, and what do we do if that materializes?"

What good looks like

• A 13-week cash flow spreadsheet that gets updated every Monday
• Receivables aging reviewed weekly, any invoice > 60 days old has a collection action
• You know your cash runway to the week, not to the month
• You have a "trigger plan", if cash drops below X, here are the three cuts we make in order

Related: P&L literacy · The three numbers · Risk management

What to do with this

• Build a 13-week rolling cash flow forecast, weekly cadence catches problems before they become crises
• Monitor DSO (days sales outstanding), every day of DSO reduction frees working capital immediately
• Keep 6-12 months of operating expenses in reserve, more for volatile businesses, less at mature scale
• Model cash flow under stress scenarios (30% revenue drop, customer payment delay), know your breaking point
• Don't confuse profit with cash, profitable businesses go bankrupt from cash crunches, not from losses