Building a cash flow forecast: a practical guide

Profitable companies go bankrupt. Not often, but more often than you'd think, and almost always for the same reason: on paper it all worked, in the bank account it didn't.

A cash flow forecast is the instrument that watches that difference. Not an accounting document, not a statutory report in cash form, but a simple question you ask yourself every month: how much cash comes in, how much goes out, and how long can I keep going if things go against me?

Most founders and SME owners only build their first forecast when it's already too late. Which is a shame, because the exercise itself isn't hard. The trap is somewhere else.

Why profit and cash are not the same

You send an invoice for €50,000. Your books show €50,000 in revenue. Your bank account shows nothing, because the client pays in 60 days. Meanwhile that month you did pay salaries, social security, rent and suppliers.

That's not an accounting issue. It's working capital, and it's the reason growth costs cash instead of generating it. The faster you grow, the more you pre-finance. Every new client is another hole to fill before they pay you back.

A cash flow forecast makes that visible before it hurts.

What has to be in it, minimum

A good forecast is rarely an impressive model. Usually it's one tab, thirteen months wide, three blocks tall.

• Inflow. Not your revenue, but the cash actually coming in. Invoices at their due date plus your clients' actual payment behaviour, not today's sales.
• Outflow. Salaries, social contributions, VAT, corporate tax, suppliers, rent, software, loan repayments. In the month they leave the account, not the month you receive the invoice.
• Balance. Starting position plus in minus out, every month. That number is your only real KPI. It's what's on your bank account.

Thirteen months isn't accidental. You want to see one full year plus the first month of the next. That way you cover year-end and the peaks around it: corporate tax, thirteenth salary, VAT year-end settlement.

Where most models go wrong

The structure isn't the problem. It's the assumptions underneath. Four classics:

1. Taking revenue as inflow

Temptation is big: sales figure in the month of sale, done. For a B2C webshop that works. For a B2B business with 30–60 day payment terms you're structurally one to two months off. And that's exactly the window where you get into trouble.

Use a DSO assumption, days sales outstanding. Look back: on average, how long does an invoice take to be paid? Take that number, not the one on your terms & conditions.

2. One scenario instead of three

A forecast with one line isn't a forecast. It's a hope. Build at least three scenarios: base, downside, and what-if-everything-goes-wrong. Not to take them all seriously, but to see which number breaks you.

Nine times out of ten your runway hangs on one parameter: one client that pays, one deal that closes, one hire you postpone. You only learn which one once you've calculated it.

3. Never looking back

A forecast you build once and then park is a dead document. The value is in the monthly ritual: actuals next to forecast, explain the delta, adjust assumptions. That takes half an hour, and it's the only moment your model gets better instead of worse.

Runway: the number every investor asks first

If you're burning cash, runway is the most important KPI. Simple: how many months can you keep going at the current burn rate?

Cash on hand divided by average monthly net outflow. Below six months it gets uncomfortable. You're already mid-fundraising while still needing to operate. Below three months you're in survival mode, which isn't a great negotiating position.

The reason investors ask that number immediately isn't curiosity. They want to know the mental state you sit down with at the table.

What belongs on your dashboard, and what doesn't

A cash flow forecast lives in a spreadsheet. The dashboard around it is minimal: three to five numbers that move monthly and you see at a glance.

• Cash position end of month, with trailing twelve-month trend.
• Runway in months, at current burn and at expected burn.
• DSO and DPO. That's where working capital moves.
• Delta between forecast and reality for the past month. Not to punish, but to learn where the model is systematically off.

Anything else doesn't belong on a cash flow dashboard. EBITDA, revenue growth, margin per product: interesting, but a different discussion. Put too much on one screen and you end up looking at nothing.

When you need help

For a start-up with one product and a handful of clients, this is a half-day exercise. As the company grows, layers appear: multiple business units, currencies, grants, deferred revenue recognition, financing structures. Then the forecast becomes maintenance.

At that point there are two options. Either it's an ad hoc project: someone builds the model, hands it over, and you run it. Or it's a recurring responsibility and you need a fractional CFO keeping the monthly rhythm. Which one applies is answered in when do you need a CFO.

Either way: start small. A simple forecast you update every month is infinitely more valuable than a complex model that surfaces once a year.

Want to spar about your cash flow?

Book a 30-minute call. I'll look at your current setup and give you an honest answer whether you're missing an hour of work or a more structural issue.