How is a cashflow forecast different from a budget?

A budget is about profit over a year: what you expect to earn and spend, usually by month, on an accruals basis, so an invoice counts the moment it is raised. A forecast is about cash and timing: it counts money on the day it moves, so the same invoice appears when the client actually pays. That single difference is why a business can be comfortably ahead of budget and still short of cash in week six. Use the budget to decide whether the year is on track, and the 13-week forecast to decide whether next month's payroll clears. Neither one substitutes for the other.

Should I put VAT in the forecast?

Yes, and gross. Money arrives in your bank including VAT and leaves including VAT, so a forecast built on net figures flatters your balance for weeks and then hits you with a bill that appears from nowhere. Show receipts and payments gross, then add the VAT payment itself as its own row in the week it leaves. Quarterly VAT is payable one calendar month and seven days after the end of the VAT period, so a quarter ending 30 June is due by 7 August. If you pay by direct debit, HMRC collect three working days after that date rather than on it — put it in the week they take it.

My income is lumpy and unpredictable. Is a forecast still worth building?

More so, not less. Lumpy income is exactly the condition a 13-week forecast exists for, because steady income needs no early warning: next month looks like this one. Split the income rows in two, confirmed and expected, and be honest about which is which. Then read the closing balance in the weeks where nothing confirmed lands. The forecast will not make your income less lumpy. It will tell you how many weeks of fixed costs you can cover without a new sale, and that number is what determines how calmly you can price, negotiate and turn down bad work.

The forecast shows a shortfall in six weeks. What do I actually do?

You have four levers, and they run in this order. Pull cash forward: chase the invoices falling due inside that window now rather than on their due date, and ask for part payment on long jobs. Push cash back: move anything discretionary — equipment, stock, a hire — beyond the low point. Cut the outflow: look at what leaves on subscription each month that nobody uses. Only then borrow. If you do need a facility, apply while the forecast still shows a healthy position rather than after the shortfall has landed, because that is when it is cheapest and quickest to arrange.

Can I just use the cashflow report in Xero or FreeAgent?

As a starting point, yes. Both will project forward from your invoices, bills and repeating transactions, which saves you the first hour of data entry. What neither can know is that one client always pays a fortnight late, that the equipment order is optional, or that you intend to take a dividend in week nine. Software forecasts from what has been recorded; a useful forecast runs on what you know. Export the projection, then move the payment dates to match how your clients behave rather than what their invoices say. That adjustment is where the value is.

Thirteen weeks or a full year — which should I be building?

Both, for different decisions, and they are not the same document. The 13-week forecast is operational: weekly, detailed, built from named invoices and known bills, used to decide what happens this month. An annual cashflow projection is directional: monthly, built out of your budget, used for the bigger questions such as whether you can afford a hire, a lease or a loan. The trouble starts when one is asked to do the other's job. An annual projection will never warn you about a tight Tuesday, and 13 weeks will never tell you whether next year's growth plan is fundable.

The Business Owner's Guide to Cashflow Forecasting

Why a forecast matters more than a P&L in the short term

Your profit and loss account tells you whether the business made money over a period. It doesn't tell you whether you'll be able to pay the wages in three weeks' time. Those are two different questions, and it's entirely possible for the answer to the first to be "yes" while the answer to the second is "not without a very uncomfortable conversation with the bank." Cash gets tied up in stock, in unpaid invoices, in a VAT bill landing at an awkward moment — none of which shows up as a problem on a profit and loss account, but all of which can genuinely stop a profitable business from being able to pay its bills.

A cashflow forecast is about timing, not profitability. It answers a narrower but more urgent question: when is money actually expected to land in the bank, and when is it expected to leave, so you can see the gaps before they open up rather than after.

Why 13 weeks

An annual budget is useful for direction. It's far too blunt an instrument for spotting a cash squeeze that might hit in four weeks' time. A rolling 13-week forecast sits in the sweet spot — long enough to see genuine trouble coming with time to act, short enough that the numbers going into it are grounded in things you actually know (confirmed invoices, known bills, contracted payroll) rather than guesswork. "Rolling" is the important word: as week one closes, you drop it off and add a new week 13 onto the end, so you always have the same three-month runway in view rather than watching the horizon shrink toward a fixed year-end.

The structure: what goes in a 13-week forecast

Keep the structure simple. A spreadsheet with one column per week and the following rows is enough to get real value from it:

Opening balance — what's actually in the bank at the start of the week. Week two's opening balance is simply week one's closing balance carried forward.
Confirmed income — invoices already raised, with a payment date you're genuinely confident about, not just the due date on the invoice.
Expected income — income you're reasonably confident about but haven't yet invoiced, or invoices where payment could slip. Keep this separate from confirmed income so you can see how much of your forecast is solid ground and how much is an assumption.
Fixed outgoings — rent, salaries, loan repayments, subscriptions, anything that happens regardless of how trading goes that week.
Variable outgoings — stock purchases, one-off supplier payments, anything tied to activity levels rather than a fixed schedule.
Closing balance — opening balance plus all income, minus all outgoings, for that week.

Add one more line beneath the closing balance: the lowest point the balance is expected to hit within that week, if payments and receipts don't land evenly across the five working days. A healthy Friday closing balance can still hide a genuinely tight Tuesday, and it's the tight Tuesday that actually causes problems.

Building it, week by week

Start with what you know for certain. List every invoice currently outstanding, and put a realistic expected payment date against each one — not the due date, the date you actually expect it based on how that client normally pays. Do the same for fixed outgoings: payroll dates, rent, loan repayments, any regular supplier terms. That's your backbone. Then layer in expected but unconfirmed income, and variable costs you can reasonably anticipate over the coming three months. Total each week, carry the closing balance forward as the next week's opening balance, and you have a working forecast.

The first version will take the longest to build. After that, updating it each week — ticking off what actually happened against what you predicted, and extending the rolling window by one more week — should take a fraction of the time, and it's this weekly discipline that makes the difference between a forecast that's useful and one that's a one-off exercise nobody looks at again.

A worked example: the week the VAT bill lands

The figures below are illustrative, but the shape of the problem is the common one. A VAT-registered design agency with a 31 December year end sits down at the end of June to build its first 13-week forecast. It has £46,000 of invoices outstanding across eleven clients, and two of those clients, worth £8,800 between them, reliably pay at around 50 days rather than the 30 printed on the invoice.

Week five is the week commencing 3 August, and three things collide around it:

The VAT for the quarter ended 30 June — £14,300 — is payable by 7 August.
Payroll of £11,200 and rent of £2,100 go out the following week.
A £6,000 equipment order has been pencilled in for that same week.

Built on invoice due dates, the forecast looks survivable. Week five opens at £16,900, takes in £12,400, pays out £14,940 — the VAT plus £640 of subscriptions — and closes at £14,360. Week six takes in £4,200, pays out £19,300, and closes at minus £740. Tight, but arguably a rounding error.

Built on how those two clients actually pay, it is a different picture. The £8,800 does not arrive in week five, so week five takes in £3,600 and closes at £5,560. Week six then closes at minus £9,540 — a real overdraft, in the same week as payroll, spotted six weeks before it happens.

Six weeks is enough time for two ordinary decisions to remove the problem completely. Chasing those two clients in the third week of July rather than the first week of August pulls the £8,800 back into week five. Moving the equipment order to week nine, once the VAT and the August payroll have cleared, releases another £6,000. Week six now closes at £5,260: no borrowing, no awkward phone call. Neither decision is clever. Both are only available to someone who could see week six from week one.

Reading it once it's built

Don't get lost in the detail of every line. Focus on the closing balance trend across the 13 weeks: is it broadly stable, climbing, or steadily eroding? A single tight week isn't necessarily a problem if it recovers the week after. A closing balance that quietly shrinks week after week for a month is telling you something important, even if no individual week looks alarming on its own — and catching that trend in week three gives you far more room to act than spotting it in week eleven.

Where forecasts usually go wrong

Two mistakes come up more than any others. The first is treating the forecast as a one-off document — built once in a burst of good intentions, then never revisited, so by week four it's more fiction than forecast. The second is being too optimistic on timing: assuming every invoice gets paid exactly on its due date, when in reality a portion always lands late. Building in a realistic buffer for late payment, based on how your clients actually behave rather than what your payment terms say they should do, makes the forecast far more useful when it actually matters.

The lever most owners under-use: getting paid sooner

Nearly every cash squeeze in a profitable business is a timing problem, and the largest piece of timing you control is how long your invoices sit unpaid. Most owners treat late payment as weather. It isn't. The Late Payment of Commercial Debts (Interest) Act 1998 gives you a statutory right, on business-to-business invoices, to charge interest at eight percentage points above the Bank of England base rate on anything paid late, plus fixed compensation towards the cost of chasing it: £40 on debts under £1,000, £70 on debts from £1,000 to £9,999.99, and £100 on debts of £10,000 or more. Where you have not agreed payment terms in writing, the default period is 30 days from the later of the goods or services being delivered and the invoice reaching the customer.

You will rarely need to charge it. The value is in the line on the invoice saying that you can, and in treating a chase as a contractual right rather than an imposition. Three habits do more for a forecast than any amount of spreadsheet refinement: put the payment date on the invoice as an actual date rather than a number of days, send the invoice the day the work finishes instead of at month end, and make the first chase automatic and early — a short note two days before the due date, not two weeks after it. There is more on this in our piece on invoicing and getting paid on time.

The same arithmetic runs in the other direction. HMRC charge late payment interest at the Bank of England base rate plus four percentage points, a margin that rose by 1.5 percentage points on 6 April 2025, so the VAT and PAYE rows in your forecast are among the most expensive lines to get wrong by a week.

Turning the forecast into decisions

A forecast that only the accountant looks at isn't doing its job. Its real value is in the hands of whoever's making the day-to-day calls — deciding whether now's the right time to place a big stock order, whether a supplier payment can wait a week, or whether it's worth chasing a slow-paying client a little harder this month rather than next. Used properly, a rolling forecast turns cash management from something you react to into something you can see coming and plan around.

How Buzz helps

This is exactly the kind of visibility we build into management accounts — regular cashflow reporting sitting alongside profit and loss, reviewed on a proper rhythm rather than built once and left to go stale. It sits alongside our cashflow and budgeting work, and if you want the reading side rather than the building side, start with how to read a cashflow forecast. If your current view of cashflow is a spreadsheet nobody's opened in months, or you'd rather have someone build and maintain the rolling forecast for you, book a free discovery call and we'll talk through what that would look like for your business.

The business owner's guide to cashflow forecasting