The 13-Week Cash Flow Forecast, Built Step by Step

A 13-week cash flow forecast sounds like something only a big company’s finance team would build. It isn’t. It’s a straightforward tool, and once you’ve built one yourself, you’ll never go back to guessing.

This is the exact method I use to build the forecast in every Cash Flow Clarity Audit. Here’s how to build your own, step by step.

What a 13-Week Cash Flow Forecast Actually Shows

Think of a 13-week cash flow forecast as a moving 13-week window into your bank account — not your revenue, not your profit, the actual cash you’ll have on hand, updated week by week as time moves forward. Each week has four pieces: the balance you started with, the cash you expect to come in, the cash you expect to go out, and the balance you end with. That ending balance becomes next week’s starting balance, and the whole thing rolls forward.

Thirteen weeks covers roughly a full quarter — close enough to be accurate, far enough out to actually plan around.

Step 1: Start With Your Actual Cash Balance

Check your actual bank balance today, across every account. Not what your accounting software says you should have — what’s really there. Every number in the forecast builds off this one, so it has to be right.

Step 2: List Your Known Weekly Cash Inflows

For each of the next 13 weeks, list the cash you actually expect to land — not when you invoiced it, when it actually clears your account. Recurring customer payments, upcoming annual contract renewals, anything else you’re confident about. If a customer typically pays 30 days late, put that payment in the week it will realistically arrive, not the week you send the invoice.

Step 3: List Your Known Weekly Cash Outflows

Same process, other direction. Payroll, rent, software subscriptions, loan payments, contractor invoices — everything leaving your account, in the week it actually leaves. Payroll weeks and rent weeks are usually predictable and worth marking clearly, since they’re often where a forecast reveals a tight week you wouldn’t have otherwise noticed.

Step 4: Roll the Balance Forward, Week by Week

Starting balance plus cash in, minus cash out, equals ending balance. That ending balance becomes the starting balance for the next week. Repeat thirteen times. It’s simple arithmetic — the value comes from doing it consistently, not from anything complicated.

Step 5: Flag Any Week That Goes Negative

This is the entire point of building the forecast in the first place. A negative week showing up in week 9 while you’re sitting comfortably in week 1 is exactly the kind of thing you want weeks of advance notice on, not a surprise.

Here’s what six weeks of a real forecast looks like, including a couple of realistic bumps — a large annual customer payment landing in week 3, and an extra contractor payment in week 5.

Week 3’s cash in jumps to $53,000 because of a $35,000 annual contract payment on top of the usual $18,000. Week 5’s cash out jumps to $37,000 because of an extra $15,000 contractor payment. Nothing here goes negative — but that’s exactly the point. This is how you’d catch it if it did, weeks before it actually happened.

Try building your own first six weeks below.

Common Mistakes When Building Your First Forecast

• Being optimistic about collection timing. If a customer has paid late twice in a row, plan around them paying late a third time. A forecast built on how people are supposed to pay isn’t a forecast — it’s a hope.
• Forgetting irregular, non-monthly expenses. Annual software renewals, insurance premiums, and quarterly estimated tax payments don’t show up every week — which makes them easy to leave out entirely, and painful when they land in a week you weren’t expecting them.
• Leaving out your own pay. If you draw a salary or owner’s draw, it belongs in the forecast like any other outflow. Founders who skip this end up with a forecast that’s quietly wrong every single week.
• Building it once and calling it done. A forecast is only as useful as it is current. One built in January and never touched again is a historical document by March, not a planning tool.

Why a 13-Week Cash Flow Forecast Matters for Decision-Making

A forecast like this doesn’t predict the future perfectly, and it isn’t supposed to. What it does is turn a vague worry into a specific, checkable number. Instead of “I think we’re okay,” you get “week 9 gets tight if that customer pays late, and here’s exactly how tight.” That’s a very different conversation to have with yourself, and a much easier one to plan around.

It’s also the difference between reacting to a hiring decision and actually deciding one. “Can I afford a new hire in week 6?” isn’t really answerable from a gut feeling — but it’s answerable in about thirty seconds once the hire’s salary is sitting in the forecast as a new weekly outflow, and you can see exactly what happens to every week after it.

It’s also worth remembering that a forecast like this is built on cash, not on your P&L. If you want the mechanics of why those two numbers can tell very different stories, I’ve written about that directly in why your P&L says profit while your bank says panic.

A Few Questions That Usually Come Up

How Is This Different From a Monthly Budget?

A budget is usually built around what you’d like to spend over a longer stretch of time. A 13-week cash flow forecast is built around what’s actually going to move through your bank account, week by week. It’s more granular and far more useful for catching a short-term problem before it arrives.

What if I Don’t Know My Exact Weekly Numbers?

Use your best realistic estimate, not a hopeful one. A forecast built on cautious assumptions is far more useful than one built on best-case guesses. You can also start with your average weekly burn as a baseline for the weeks you’re less certain about.

How Often Should I Update My 13-Week Cash Flow Forecast?

Weekly, ideally. Update it with what actually happened, then let it roll forward another week. A forecast you build once and never touch again loses its accuracy fast — the value comes from keeping it current.