A free 13-week cash flow forecast you can fill in right here.

This is the same instrument we build inside the companies we work with, stripped to its essentials and put here for you to use free. No email, no sign-up, and your figures never leave your browser: they are stored on your own device and we never see them.

It loads with a made-up company so you can see how it works. Replace the numbers with yours and the table recalculates itself.

What it is telling you

The example it loads with is not random. It is the case we run into most: a company that looks healthy for eleven weeks and runs out of cash in week twelve.

Look at the chart. For almost the whole quarter the balance sits between $27,000 and $74,000. Nothing alarming in a month-end close. And yet, in the week where payroll lands alongside a quarter of VAT already paid, the balance crosses zero.

A monthly forecast would not have seen this. It would have shown a month where money in and money out roughly balance. The specific week where payments collide only appears when you look week by week, and that is the entire reason for the thirteen columns.

How to use it with your own data

Start with the real balance, not the accounting one. What is in the bank today, across all accounts. Not the balance according to your books, which will be weeks behind.

For receipts, enter the date you will actually be paid. Not the agreed due date. If a customer is on 60-day terms and consistently pays at 85, enter the week that corresponds to 85. That is the difference between a forecast and a wish list.

Include confirmed orders you have not invoiced yet. That is half of any serious cash forecast, and it is exactly what gets left out when someone works only from issued invoices.

Do not forget what does not happen every month. Quarterly VAT, bonuses, corporation tax, the annual insurance premium. These are what break forecasts, precisely because they are not part of the monthly rhythm.

The two scenarios we built in

The sliders ask the two questions that change the answer most.

"What if customers pay later." It is the most likely scenario of all and almost nobody models it. Try it on the example and you will see something that is not obvious.

At 14 days of delay, the week-12 hole goes from $600 to $18,600. Same week, thirty times deeper. But at 21 days something else happens: the break jumps from week 12 to week 4.

In other words, this does not degrade gradually: there is a cliff. While the delay stays small, the problem is the same one, only worse. Past a certain point, cash breaks two months earlier and all your room to manoeuvre disappears. Finding where that cliff sits in your own company is probably the most useful thing you can get out of this tool in ten minutes.

"What if sales drop." This one affects only the money coming in, which is how it works in reality — fixed costs do not fall with sales, and that is why a 15% drop hurts more than it looks like it should.

What to do with what you see

If a week comes out red, you have four levers and they are best used in this order: pull a collection forward (call the large customer early, offer an early-payment discount), push a payment back (talk to the supplier before it falls due, not after), draw on your credit line — that is what it is for, and it is far cheaper to negotiate weeks ahead than on the day you need it — and finally cut costs, which is slowest and therefore last.

What matters is that all four need time. Which is why the value of this table is not its precision: it is seeing the problem with six weeks of room instead of six days.

And then the part no tool automates

Refreshing it every Monday. Drop the completed week, add a new week 13, and compare what you forecast against what actually happened.

That comparison is where the learning is: if week after week you collect less than forecast, it is not bad luck — your collection assumptions are optimistic and need correcting. A forecast that is not refreshed becomes a stale photograph within a month.