A cash flow forecast is a pipeline, not a spreadsheet you rebuild every month
What a cash flow forecast is
A cash flow forecast projects the timing of money entering and leaving your bank account, so you can see a shortfall before it arrives rather than after it has already caught you out. It works on a running balance: you start with the cash you hold today, add what you expect to receive, subtract what you expect to pay, and carry the result forward week by week or month by month. This page goes deep on forecasting alone. The mechanics of getting invoices out and cash collected sit within your wider financial operations.
Every forecast is built from five moving parts:
- Opening balance: the cash you hold at the start of the period.
- Cash in: receipts you expect, dated when the money actually lands.
- Cash out: payments you expect, dated when they actually leave.
- Net cash flow: cash in minus cash out for the period.
- Closing balance: opening balance plus net cash flow, which becomes next period's opening balance.
Forecast, budget or projection?
These three terms get used interchangeably, and that muddle is where a lot of financial confusion starts. They answer different questions. A budget is what you committed to, a forecast is what you now expect, and a projection is a modelled scenario you are testing.
| Term | What it is |
|---|---|
| Budget | The plan you commit to at the start of the year. |
| Forecast | Your best current expectation of what will actually happen. |
| Projection | A modelled what-if scenario, such as losing a key client. |
Keep that distinction in mind, because a business can be profitable on paper and still run short of cash, and the forecast is what catches that gap before it bites.
Why the hand-built forecast keeps failing
Most owners already keep a cash flow forecast. It usually lives in Microsoft Excel or Google Sheets, gets rebuilt from scratch each month, and is re-keyed by hand from the accounting system, so it is stale within days. By week three it has become another example of when a spreadsheet quietly stops being trusted, and it gets set aside. None of that is a failure of discipline: the same errors catch everyone, from over-optimistic revenue and invoice due dates treated as payment dates to ignored seasonality and profit mistaken for cash.
How can a profitable business run out of cash?
The gap between the two lives in timing. Your profit-and-loss account records a sale the moment you raise the invoice, so a strong month reads as a strong month. Your bank account only moves when the money actually lands, and that can be weeks later. When you are growing, you fund the work (materials, wages, overheads) before the payment for it ever arrives.
The trap: Profit is an accounting opinion. Cash is a fact. A sale sitting on the profit-and-loss account is a promise, and a promise does not pay wages.
Picture a month where you invoice, say, £40,000 of genuinely profitable work. Your terms are 30 days, but customers settle at 45 to 50. You have already paid suppliers and run payroll for that job, and a VAT quarter falls in the same window. On paper you made a profit. In the bank, the weeks look like this.
| Week | Cash in | Cash out | Bank balance |
|---|---|---|---|
| Start | - | - | £20,000 |
| 1 | £3,000 | £11,000 (materials) | £12,000 |
| 2 | £2,000 | £9,000 (payroll) | £5,000 |
| 3 | £1,000 | £4,500 (overheads) | £1,500 |
| 4 | £2,000 | £6,000 (VAT) | -£2,500 |
The balance falls below zero while the job itself was never unprofitable. That gap is your cash conversion cycle: debtor days (how long customers take to pay you), creditor days (how long you take to pay suppliers), and the stock sitting on your shelves waiting to sell. Cash is trapped in the first and the last. Collecting sooner shortens debtor days and frees it, holding less stock frees more, and paying suppliers a little later widens the gap in your favour. A cash flow forecast puts that timing on the table before the bank balance, not the profit figure, decides whether you can make payroll.
The simple build: your first forecast
You can put a working forecast together this afternoon in a single spreadsheet. It does not need to be clever. It needs to be dated, honest about timing, and updated as real numbers land. Follow these steps in order and you will have something you can act on today.
Pick a period and granularity
Go weekly if cash is tight, monthly if it is comfortable. Set the horizon to twelve columns so you can see far enough ahead to react.
Start with today's real balance
Open with the actual figure in the bank right now, not the accounting balance. This is the only number you know for certain, so everything else builds off it.
List dated inflows
Enter each receipt in the period the money will really arrive, based on when customers pay, not the invoice date. Late payers get dated late.
List dated outflows
Add supplier bills, payroll, rent and tax against the weeks they leave the account. Pulling this from your supplier bills and payment runs keeps the outflow half accurate rather than guessed.
Run the balance forward
For each period, take the opening balance, add inflows, subtract outflows, and carry the closing figure into the next column. Then review and correct as reality comes in.
Worked small: say you open the month at £8,000. Week one brings in £6,000 and pays out £9,000, closing at £5,000. Weeks two to four net roughly level, and a £4,000 tax payment in week four closes the month near £1,000. The dip in week one is the moment worth seeing early.
This is the direct method: actual receipts and payments, dated as they happen, which is precisely what makes a forecast buildable from your operational data rather than reverse-engineered from profit figures the way the indirect method works.
A forecast as a live pipeline, not a monthly rebuild
Because the direct method builds the forecast from money actually moving, every line already traces back to a system the business runs. So nobody needs to re-key anything. The forecast stops being a monthly rebuild and becomes a pipeline: each input arrives from its source, updates on its own schedule, and never goes stale between refreshes.
Sales ledger (receivables)
Every outstanding invoice: who owes what, and when each one falls due. The raw material of the receipts side.
Purchase ledger (payables)
Bills accepted but not yet paid, with each supplier's terms setting the timing of money going out.
Payroll
Wages, PAYE and pension contributions, landing on fixed dates and rarely flexing in the short term.
HMRC tax calendar
VAT quarters, PAYE and corporation tax deadlines: large, predictable outflows a forecast should never be surprised by.
Order book
Committed and pipeline orders that become future invoices, so inflows reflect real demand rather than guesswork.
None of this needs a new home. The data already sits in Xero or Sage, with day-to-day capture through a bank feed (open banking such as TrueLayer or Plaid) and receipt tools like Dext. What we build is a light layer over that existing accounting system, not another subscription. Dedicated tools such as Float, Fathom and Agicap already sell live-data forecasting, so the idea is not the point. The point is the architecture underneath: knowing what feeds what, whichever package holds it.
Forecasting when customers actually pay
An invoice is rarely paid on the day it falls due. If a customer reliably pays in thirty days, forecasting that receipt at fourteen because the terms say so overstates next week's balance. Realistic collections timing is table stakes. The mechanism that makes it hold: derive each customer's or segment's real payment pattern from their history in the sales ledger, then feed that timing in automatically, so the receipts line reflects behaviour rather than hope. A payer who always takes forty-five days is forecast at forty-five. This is the same view of how each customer actually pays that drives day-to-day credit control, pointed forward instead of chased backward.
UK tax on the calendar: VAT, PAYE and corporation tax
These outflows are the most predictable money leaving your account all year, and they are also the ones that blindside owners most often. They arrive in lumps, they do not care what your sales pipeline looks like that week, and each has a fixed date set by HMRC. A forecast earns its keep by absorbing them well ahead of time. Where your bookkeeping runs on Making Tax Digital data, the numbers are already clean enough for the forecast to place each payment on the right day automatically, rather than you keying dates by hand.
| Payment | When it falls due | How often |
|---|---|---|
| VAT | One calendar month and seven days after the VAT quarter ends | Quarterly (most filers) |
| PAYE and NIC | By the 22nd of the following month if paying electronically (the 19th by post) | Monthly |
| Corporation tax | Nine months and one day after the accounting period ends | Annually |
| Payments on account | 31 January and 31 July (sole traders and partners) | Twice yearly |
| Loan and finance repayments | Your agreed instalment date | Monthly, usually |
None of these should ever be a surprise. A forecast that carries them all is a forecast you can trust.
The 13-week cash flow forecast
A 13-week cash flow forecast plots your money week by week across roughly a quarter. Thirteen weeks is long enough to act on a shortfall and short enough to stay honest: far enough out to reroute a payment or chase an invoice, near enough that the numbers still describe something real. Because it is built on the direct method, tracking cash actually landing and leaving, it can be assembled straight from the live pipeline you already feed.
Restructuring and turnaround firms tend to own this term and frame the 13 week cash flow as a distress signal, a tool you reach for when the bank is nervous. It is not only that. For a healthy but tight business that is growing, it is the natural horizon: far enough ahead to act, without forecasting so far out that the numbers become fiction.
The rolling forecast
A rolling forecast is a moving window. Each period you drop the oldest week and add a fresh one at the far end, so the forecast never runs out and re-forecasting becomes continuous rather than a once-a-year event. Update it monthly as standard, and weekly when cash is tight. That contrasts sharply with a static annual budget, which is stale by February and never catches up.
| Approach | Static annual budget | Rolling forecast |
|---|---|---|
| Horizon | Fixed to year-end, shrinking as months pass | Always the same distance ahead |
| When it goes stale | Within weeks of being set | Never, because the far end is refreshed |
| Re-forecasting | An annual event, often skipped | Continuous, monthly or weekly |
Honest uncertainty: a range that narrows, not a single number
A forecast is not a promise of one number. It is a range that gets narrower as real data arrives. Instead of defending a single confident line, you model what-ifs and watch each one flex the cash position. That is scenario analysis, with sensitivity analysis underneath it, and live toggles you can pull up in a meeting beat three static best, worst, and likely columns printed on a page.
The big invoice slips 30 days
Shows how deep the cash dips before payment lands, and whether payroll survives the gap.
You lose the biggest customer
Shows how fast your headroom erodes when the largest inflow stops, and how long the runway holds.
A seasonal dip lands
Shows the quiet months early, so you build a buffer rather than scramble.
The range tightens through a variance loop. Each cycle you compare the forecast against what actually landed, then feed the gap back so the next projection is sharper. That loop depends on clean actuals: records that agree across bank, invoices, and ledger. The aim is knowing how much headroom you hold and where it could break, not a single number to cross your fingers over.
Software or spreadsheet: the honest choice
Most cash flow forecasting software comparisons are affiliate round-ups pushing whichever tool pays the best commission. Here is the straight version. There is no single right tool, only a right fit for how your cash actually behaves. The question is not which product tops a list, it is which of three honest positions matches your situation today.
| When it fits | The right choice |
|---|---|
| Cash is simple, one person owns the forecast and keeps it current | A spreadsheet |
| Re-keying is the real problem and the numbers already sit in Xero or Sage | A light custom layer |
| You need rich scenario modelling out of the box and can absorb another subscription | A dedicated tool (Float, Fathom) |
When re-keying is the bottleneck and the data already lives in your accounting system, whether that is Xero, Sage, QuickBooks or FreeAgent, the light pipeline over it is what we build.
Common questions
Can a profitable business really run out of cash?
Yes. It is a timing problem, not a profit problem. Cash gets tied up in unpaid invoices and stock while VAT, PAYE and payroll still fall due on fixed dates.
How accurate is a cash flow forecast?
It is a range that narrows as real data arrives, not a single guaranteed figure. The next few weeks are reliable; distant weeks are looser. Scenario toggles show where the plan could break.
Do I need software, or is a spreadsheet enough?
A spreadsheet is fine when cash is simple and one person keeps it current. When re-keying becomes the bottleneck and your numbers already live in Xero or Sage, a light layer over them wins.
What is the best cash flow forecasting software for a small UK business?
The best tool is the one that stays current with the least manual work. Often that is a light layer over your existing accounting system, sometimes a dedicated tool like Float or Fathom.
See the shortfall coming
The end state is a plain one. You see the shortfall eight weeks out, while you still have room to chase an invoice, hold back an order or arrange cover, rather than the morning payroll fails to clear. A cash flow forecast is a timing tool, not a profit report, and the version that survives is a live view your existing systems feed, updated as a rolling window. The honest uncertainty is the point: a range that narrows and a few scenario toggles turn it from a guess into something you can act on.
See where your cash will be in eight weeks
No template to download. We wire a live forecast over your Xero or Sage and keep it current, so the number is there when you need it.
Talk to us about your forecast →