How to Create a Cash Flow Forecast for Small Business Success

You open your banking app on a Tuesday morning. There's €8,400 in the account. On Friday, payroll goes out: €11,200. And you only find out about the gap when the payment bounces.

That morning happened to me in 2023. Not because the business was failing — revenue that quarter was actually up. It happened because I was running the company off my bank balance, which is a rearview mirror, not a windshield. A cash flow forecast is the windshield. And building one for a small business is genuinely simpler than most finance people make it sound. It also takes less time than you'd guess: my first working version took me one Sunday afternoon.

What a cash flow forecast actually is (and what it isn't)

A cash flow forecast is a week-by-week or month-by-month table of money coming in and money going out, based on when the money actually moves, not when the invoice is dated. That distinction is the whole point, and it's where most small business owners get caught.

You send an invoice on the 3rd. Your accounting software logs revenue on the 3rd. Your bank account sees nothing until the 27th — or the 12th of next month, if the client pays late. A forecast tracks the 27th.

Profit is not cash, and that trips people up

I've watched profitable businesses run out of money. It's not rare, it's routine. You can show a healthy margin on paper while the actual bank balance dips below what you need for rent, because your biggest customer decided to pay in 60 days instead of 30.

The forecast exists to answer one question: will I have enough money in the account on the day each bill lands? Everything else is decoration.

How to create a cash flow forecast for a small business in five moves

Here's the sequence I use, refined over several years of getting it wrong first. No software required for version one — a spreadsheet is fine, and honestly better for understanding the mechanics.

Step 1: choose your period and match it to your cash cycle

Start with a 12-month projection for planning, then zoom into weekly detail for the next 8–13 weeks. Weekly granularity is where the useful information lives for a small business, because rent, payroll and supplier payments all cluster around specific dates.

On horizon length: your forecast should stretch at least as long as the time it takes money to travel from "customer owes you" to "money is in the bank." If your customers take 45 days to pay and your suppliers give you 15 days to pay them, a 30-day forecast is structurally blind. I made exactly that mistake in year one.

Step 2: list every cash inflow with a realistic date

Not the invoice date. The date you expect the money to land. For each one you need:

• Customer name and invoice amount
• Expected payment date based on how that specific client actually pays you (check your bank history — some pay in 12 days, some in 55)
• Recurring income, like retainers or subscriptions, listed separately because it's more reliable
• Any other inflows: deposits, loan drawdowns, tax refunds, asset sales

The trap here is using your nominal payment terms. If your contract says 30 days but the client historically pays in 47, forecast 47. I keep a simple note per client — "Pays late, always" — and it has saved me twice.

Step 3: list every cash outflow, including the ones you'd rather forget

This is where small business forecasts usually fail. People include rent and salaries, then forget:

• VAT or sales tax — quarterly or monthly, and it stings
• Corporation tax or income tax instalments
• Owner's drawings — your own salary, which is a real outflow
• Loan repayments and interest
• Annual costs, spread across the year: insurance, software renewals, accounting fees
• Seasonal spikes: stock build-up before your busy period

My first forecast omitted the quarterly VAT payment entirely. It looked fine for three months, then showed a €6,000 hole that wasn't real — it was money I owed and had forgotten to plan for.

Step 4: calculate the running balance

Add a row for the opening balance, then for each period: opening balance + inflows − outflows = closing balance, which becomes the next period's opening balance. This running total is the only number that matters. The individual lines are just ingredients.

Watch the minimum across all periods. If the closing balance dips negative in week seven, you have six weeks to do something about it. That's the entire value of the exercise.

Step 5: build two scenarios, not one

A single forecast is a guess wearing a suit. Build a base case and a pessimistic case where your two biggest clients pay a month late and one sale doesn't close. The pessimistic case isn't pessimism — it's the version you can survive.

I run three on my own numbers: base, slow-payment, and "lost the biggest client." That third one forced me to keep a cash buffer I'd otherwise have spent. It hurt at the time. Then a client left in 2024 and nothing broke.

Spreadsheet, accounting software or bank tool: which one to use

You have three realistic options. I've used all three, and my honest opinion is that most small businesses should start with the spreadsheet and only move when it becomes genuinely painful.

Option	Best for	Weakness	Typical cost
Spreadsheet (Excel, Google Sheets)	Getting your head around the mechanics; under ~50 transactions a month	Manual entry, breaks easily, easy to forget to update	Free to low
Accounting software module	Businesses already invoicing through the platform; auto-pulls receivables	Forecasts inherit your invoicing data, so it's only as good as your payment-date assumptions	Included in most mid-tier plans
Dedicated cash flow tool or bank-embedded forecast	Multi-account businesses, several currencies, real-time bank feeds	Cost, setup time, and you still must interpret the output	Monthly subscription

Start with the spreadsheet. Not because it's better, but because building it by hand teaches you which lines actually move your cash — a lesson no automated tool will give you. Once your forecast format is stable and updating it takes more than 20 minutes a week, switch.

A cash flow forecast example you can copy

Here's a simplified version of what a small design studio's forecast looks like over four weeks. Numbers are illustrative but the structure is exactly what I use.

Item	Week 1	Week 2	Week 3	Week 4
Opening balance	8,400	9,100	1,900	6,400
Client payments received	4,000	0	9,500	2,000
Salaries	−3,300	0	−3,300	0
Rent and utilities	0	−1,200	0	0
Suppliers and software	0	−6,000	−1,700	−900
VAT payment	0	0	0	0
Closing balance	9,100	1,900	6,400	7,500

Week 2 is the story. A €6,000 supplier payment lands on the same week as rent, and the balance drops to €1,900. That's survivable but tight. If payroll had fallen in week 2 instead, the studio would be negative — and they'd have two weeks' warning to chase an invoice or negotiate terms.

That's the whole point of the exercise in one row.

The mistakes that make forecasts useless

I've made most of these, so this list is worn in.

Assuming your invoices get paid on time

The single biggest source of error. Track each client's real payment behaviour over the last six months and use the average actual delay, not the contractual one. Most small businesses I've compared notes with find their real collection time is roughly 15–25 days longer than the terms they quote.

Forgetting tax and owner drawings

Both are real cash outflows. Excluding them makes the forecast look better than reality and removes the one benefit it has: warning you.

Updating it once a quarter

A forecast built in January and never touched is a fiction by March. Update weekly — 10 minutes is enough once the structure exists. Note the variance between what you predicted and what actually happened; over a few months, your predictions get noticeably sharper.

Running only one scenario

The base case is what you hope happens. The pessimistic case is what you plan against. If you only build one, you've written a wish list.

The one habit that matters more than the format

The template you choose barely matters. The spreadsheet layout is not the problem. What separates a forecast that saves your business from a document nobody opens is check-in frequency — and specifically, the small discipline of comparing last week's prediction to this week's reality and writing down the gap.

Do that for three months and you'll start to see your business the way a pilot sees weather: not as a surprise, but as a pattern arriving. The €11,200 payroll that once caught you out becomes a date circled on a calendar, and you have already decided what to do about it.

The first version of your forecast will be wrong. Mine was wrong for months. Build it anyway — a wrong forecast you update is worth far more than a perfect one you never open.