13-Week Cash Flow Forecast for Startups: How to Build One (With Example)

What is a 13-week cash flow forecast?

A 13-week cash flow forecast is a weekly table of the money you expect to receive and pay over the next quarter. Each column is one week, each row is a type of cash movement, and the bottom line is the cash balance at the end of every week. It covers 13 weeks because that is one quarter.

It is a cash tool, not an accounting tool. A sale only counts in the week the money lands in your bank account, and a cost only counts in the week it leaves. That is the single biggest difference from a profit and loss statement, and the reason a company can look profitable on paper and still miss payroll.

The forecast is rolling. Every week you replace the week that just ended with what actually happened and add a new week at the far end, so you always look 13 weeks ahead.

Why 13 weeks and not a full-year model

A long financial model answers the question "what could this company become?" A 13-week forecast answers a different one: "will we have enough cash on a specific Friday?" Beyond about three months, weekly guesses about collections and vendor payments are mostly noise. Inside three months, you know most of your payroll, rent and contracts, so the numbers are firm enough to act on.

Short horizons also make decisions concrete. If the forecast shows cash dipping below your safety level in week 9, you have eight weeks to chase an invoice, delay a hire or start a bridge conversation. Monthly runway math cannot show that timing, because it averages good and bad weeks together.

Think of the two tools as a team. The financial model sets direction, and the 13-week forecast checks that you survive the road to it.

How to build one, step by step

Start with the opening cash balance from your bank account, not from your books. Then list cash coming in, cash going out and the net change for each week. Closing cash for one week becomes opening cash for the next. A spreadsheet with 13 week columns is enough; you do not need special software.

Fill in the certain rows first: payroll dates, rent, subscriptions, loan payments and tax dates. These are known, and they usually make up most of your outflow. Then add the uncertain rows, mainly customer payments. For each expected receipt, put the amount in the week you think it will be paid, then challenge yourself on that date.

Finish with three summary lines: net cash change, closing cash and the lowest balance in the 13 weeks. That last number is the one to watch. A healthy average can hide a single week where you cannot pay people.

A worked example with simple numbers

Take an illustrative startup with $120,000 in the bank. Over the next 13 weeks it expects $78,000 of customer payments. It pays $90,000 in payroll, $12,000 for software and hosting, $9,000 for rent and office costs and $6,000 for contractors. Cash out is $117,000, so net change is $78,000 − $117,000 = −$39,000.

Closing cash is $120,000 − $39,000 = $81,000. Average net burn is $39,000 ÷ 13 = $3,000 a week, so the remaining runway at that pace is $81,000 ÷ $3,000 = 27 weeks. These are made-up figures to show the mechanics, not a benchmark.

Now stress it. Suppose an $18,000 invoice you expected in week 5 slips beyond week 13. Closing cash falls to $63,000, average burn rises to about $4,385 a week and runway drops to roughly 14 weeks. One late customer cut the weeks of cash in half, and a monthly average would not have shown it coming.

Forecasting cash in: collections, not invoices

The most common error is forecasting when you send an invoice instead of when you are paid. If a customer's terms are net 30 and they usually pay at day 45, the money belongs in week 7 after the invoice, not week 1. Use your real payment history, which our guide to days sales outstanding explains, rather than the terms on paper.

Treat each source differently. Subscription payments by card are predictable and land within days. Annual prepayments arrive in a lump, so check the deferred revenue picture before you celebrate. Large B2B and government contracts are the least predictable, with approvals and payment runs that can slip by weeks.

A practical rule is to forecast conservatively. Move uncertain receipts later, count only signed deals, and keep hoped-for deals in a separate list that does not feed the balance. Upside is easy to add later; a surprise gap is not.

Cash out rows and your minimum cash level

List outflows by the date they leave the account. Payroll is often twice a month, taxes and benefits come on their own dates, annual software renewals can land in a single week, and card bills trail the spending by a month. Put each one in its real week instead of spreading it evenly, because the uneven weeks are what the forecast exists to catch.

Then set a minimum cash level, a floor you do not want to go below. Many teams choose a few weeks of payroll plus fixed costs, but the right number depends on how predictable your income is. Mark every week where the forecast breaks that floor. Each mark is a decision to make early, such as delaying a hire, negotiating terms or starting a fundraise.

Remember that gross burn and net burn tell different stories. Our guide on the two explains why the forecast should show both the money out and the money in separately.

A weekly routine and common mistakes

Pick one weekday and keep a fixed 30-minute routine. First, replace last week's forecast with actual bank movements. Second, note the biggest differences and ask why they happened. Third, move the next 13 weeks forward, adjust dates for late payers and look at the lowest balance again. Over a few months, you will learn how reliably you forecast.

The common mistakes are predictable. Teams forecast profit instead of cash, forget one-time payments such as taxes or annual licences, and never compare forecast to actuals, so the model stays wrong. Another mistake is building a perfect forecast once and never updating it. A rough forecast updated weekly beats a beautiful one that is a month old.

Keep the sheet simple enough that two people can read it in five minutes. If you need a long explanation, investors, advisors and co-founders will not use it. Our guide on runway extension shows what to do once the forecast flags a problem.

Practice the cash discipline in Founder Runway

In Founder Runway, a run lasts 20 turns and every decision has four options. You pick a B2B, B2C or B2G business model and start at Pre-Seed, Seed, Pre-Series A or Series A. It is a learning game, not financial advice, but it lets you feel how a slow customer payment or an early hire changes your cash.

A run can end in failure, a promising outcome, EBITDA positive or high-value exit potential. Replay with a different model and watch which turns put your cash under pressure.