The 13-week cash flow forecast, explained for owners

If you run a company between $1M and $10M and you only ever build one financial tool, build this one. It takes a day to set up, twenty minutes a week to maintain, and it is the difference between managing cash and reacting to it.

What it actually is

A 13-week cash flow forecast is a simple grid. Thirteen columns, one per week. Down the side: cash coming in, cash going out, and where the bank balance lands at the end of each week. That is the whole thing. It is not a budget and it is not a P&L. It ignores accruals, depreciation and every accounting convention that puts distance between a transaction and the money moving.

Why thirteen weeks

One quarter is the horizon where you can still do something about what you see. Four weeks is too short — by the time a problem appears, your options are borrowing or not paying someone. Twelve months is too long to be accurate at a weekly level. Thirteen weeks is far enough out to give you room to act, close enough in that the numbers are real.

Building your first one

1. Start from the bank, not the books

Open your last 12 weeks of bank statements. This is your baseline reality: what actually hit the account and when. Your accounting system tells you when you invoiced. The bank tells you when you got paid. For cash forecasting, only the second one matters.

2. Lay out cash in

Collections from existing receivables, week by week, based on when that customer actually pays — not their stated terms. New sales you can reasonably expect. Anything else: a tax refund, a draw on a line of credit, a deposit.

3. Lay out cash out

Payroll and payroll taxes on their real dates. Rent. Loan payments. Recurring software and insurance. Then supplier payments based on your actual pattern. Then the lumpy ones people forget: quarterly tax deposits, annual insurance renewals, equipment, owner distributions.

4. Add the ending balance line

Opening balance, plus in, minus out, equals closing — which becomes next week’s opening. That row is the whole point. It is where you see the week you go negative, six weeks before you get there.

Using it

Every Monday, update two things: the actual closing balance from last week, and anything you now know that you did not know before — a customer who slipped, a job that closed, an invoice that will land late. Roll the window forward one week. Twenty minutes.

Then ask the only question that matters: what is the lowest point in the next thirteen weeks, and how close is it to zero? That number, tracked weekly, changes how you make decisions. Hiring, equipment, taking on a job with 60-day terms — all of it gets easier when you can see the trough before you commit.

The three mistakes

• Forecasting invoices instead of collections. If your average customer pays in 47 days, forecast 47 days, not the 30 on the invoice.
• Leaving out the lumpy stuff. Quarterly taxes and annual renewals are what turn a comfortable forecast into a surprise.
• Building it once. A forecast you do not update is a historical document. The value is entirely in the weekly rhythm.

When to get help

If you have more than a handful of customers, work in progress, inventory, or debt with covenants, the model gets complicated fast — and a forecast that is wrong is worse than no forecast, because you will trust it. That is the point where it is worth having someone build it properly and run it with you.