Short answerA 13-week cash flow forecast projects actual cash in and out of your bank account, week by week, over one rolling quarter. Built on the direct method — receipts minus disbursements — it shows your true ending cash balance each week and typically gives six to eight weeks of warning before a shortfall.
Your monthly P&L can say you're fine while your bank account dies on the 15th. Payroll lands on the 15th and the 31st; your biggest customer pays on the 28th. Monthly numbers average that gap away — a 13-week cash flow forecast is how you see it coming. Per Graphite Financial's guide for startups, rolling the forecast forward weekly gives you six to eight weeks of warning before a shortfall hits the account. Here's the direct-method build, with a template you can copy today.
What is a 13-week cash flow forecast and how do you build one?
It's a weekly, direct-method projection of cash receipts and cash disbursements over the next 13 weeks — one full quarter. "Direct method" means you forecast physical bank movements (customer payments, payroll runs, vendor checks), not accounting entries. The tool comes out of corporate restructuring, as Wall Street Prep notes — it's what advisors build first when a company is in trouble, precisely because it answers the only question that matters under pressure: how much cash will actually be in the account, each week?
You don't need to be in trouble to use it. You need it because startup cash dies of timing, not of averages.
The build, in five steps:
- Set up 13 weekly columns with beginning cash, receipts, disbursements, net cash flow, and ending cash. Ending cash in week 1 becomes beginning cash in week 2.
- Forecast receipts by customer. Take open invoices and expected sales, then place each payment in the week it will actually arrive — based on how each customer really pays, not their contractual terms. The customer on net-30 who pays in 45 days goes in the week 45 days out.
- Forecast disbursements by commitment. Payroll in its exact weeks, rent on the 1st, quarterly taxes and insurance in their true weeks, vendors per their terms, credit card settlement dates. Precision on payroll weeks matters most — it's the biggest and least movable line.
- Compute net flow and ending cash weekly. Flag any week where ending cash drops below your minimum comfort level (one payroll cycle is a common floor).
- Reconcile every week. Each Monday, replace last week's forecast with what actually happened, note the variance, adjust the assumptions it invalidates, and add week 14. The model is a living process, not a file.
The template: copy this structure
| Line | Wk 1 | Wk 2 | … | Wk 13 |
|---|---|---|---|---|
| Beginning cash | 480,000 | 438,500 | … | … |
| Receipts | ||||
| Customer collections (by name, top 10) | 65,000 | 20,000 | … | … |
| Other inflows (tax refunds, interest) | 0 | 1,500 | … | … |
| Total receipts | 65,000 | 21,500 | … | … |
| Disbursements | ||||
| Payroll + payroll taxes | 82,000 | 0 | … | … |
| Rent & facilities | 12,000 | 0 | … | … |
| Software & infrastructure | 4,500 | 3,000 | … | … |
| Marketing / ad spend | 6,000 | 6,000 | … | … |
| Insurance, taxes (in their true weeks) | 0 | 0 | … | … |
| One-offs (legal, equipment, deposits) | 2,000 | 0 | … | … |
| Total disbursements | 106,500 | 9,000 | … | … |
| Net cash flow | −41,500 | +12,500 | … | … |
| Ending cash | 438,500 | 451,000 | … | … |
| Weeks below minimum? | no | no | … | flag |
Numbers above are illustrative — the structure is the template. Two build rules from Dwight Funding's practitioner guide worth adopting: name your top customers individually on the receipts side (anonymous "revenue" lines hide your collection risk), and derive assumptions from your last 12 months of actual payment behavior, not from what contracts say.
Why weekly beats monthly when cash is tight
A monthly forecast told you that March nets out to −$41K. Fine. What it didn't tell you: in week 2 of March, ending cash dips to $9K because payroll lands twelve days before your largest receivable. You'd have survived on paper and bounced payroll in reality.
That's the original angle most guides skip: the direct method's real product isn't the forecast — it's the argument with your own optimism. Placing each inflow in the week it truly arrives forces you to confront that "net-30" is a fiction for half your customers. Founders who run this discover their true average collection lag within three weekly reconciliation cycles, and that number quietly fixes their runway math too.
The 13-week view also changes behavior upstream. When you can see the week-9 dip today, you have options that don't exist in week 8: chase collections early, offer a 2% discount for prepayment, shift a vendor payment, delay a hire, or draw a credit line while you still qualify. Six to eight weeks of warning is the difference between managing and begging.
How it fits your other forecasts
The 13-week model doesn't replace your 3-statement model — they answer different questions on different clocks. The monthly integrated model, part of the financial model VCs actually read, handles the 12–36 month story: hiring, milestones, the raise. The 13-week model handles survival mechanics. Investors increasingly expect both; a weekly cash process signals operational discipline the way clean books signal accounting discipline. And your gross vs net burn figures get sharper when they're computed from reconciled weekly actuals instead of P&L approximations.
What this means for you
Decision rule: if your runway is under 12 months, or a single customer is more than 20% of collections, you should be running a 13-week forecast — starting this week, in a plain spreadsheet, even if the first version is rough. The first draft takes an afternoon; the weekly update takes under an hour; the first time it catches a payroll-week dip, it pays for every hour you'll ever spend on it.
Frequently asked questions
Why 13 weeks and not 12 or 26?
Thirteen weeks is exactly one quarter, which aligns the forecast with board reporting and covers a full cycle of monthly and quarterly payments — rent, payroll runs, quarterly taxes, insurance. It's long enough to act on a warning and short enough that weekly estimates stay honest rather than becoming guesses.
What's the difference between the direct and indirect cash flow method?
The direct method lists physical cash movements — customer payments in, payroll and vendor payments out — by the week they hit the bank. The indirect method starts from net income and adjusts for non-cash items and working-capital changes. Direct is more precise short-term, which is why the 13-week model uses it.
How accurate should a 13-week forecast be?
Aim for ending-cash variance within about 5% in weeks one to four; further weeks can be looser. The point isn't perfection — it's that weekly reconciliation surfaces your misses fast, so your assumptions about collection timing and spend improve every single week you run the process.
Who should own the 13-week forecast in a startup?
Someone who sees both the bank account and the commitments — usually the founder at pre-seed, then a finance lead or fractional CFO. The weekly update takes 30–60 minutes once the template exists. What matters is a single owner and a fixed weekly rhythm, not the job title.
Sources
- Graphite Financial — What is a 13-Week Cash Flow (TWCF) Forecast: Guide for Startups. https://www.graphitefinancial.com/blog/why-you-need-13-week-cash-flow-forecast (accessed July 2026)
- Wall Street Prep — 13-Week Cash Flow Model (TWCF): Template Example. https://www.wallstreetprep.com/knowledge/demystifying-the-13-week-cash-flow-model-in-excel/ (accessed July 2026)
- Dwight Funding — 13-Week Cash Flow Model: Template and Guide. https://dwightfunding.com/the-practical-guide-to-establishing-a-13-week-cash-flow-model/ (accessed July 2026)
