A 13-week cash flow forecast is a week-by-week projection of the actual money moving in and out of your bank account over the next quarter. Cash flow forecasting for small business owners exists to answer one blunt question: on any given Friday in the next three months, will there be enough in the account to pay staff, suppliers and HMRC? You build it by starting with today’s bank balance, adding the cash you realistically expect to collect each week, subtracting every payment you know is due, and carrying the closing figure forward as next week’s opening figure.
The reason it works is that it deals in dates and pounds, not profit. A sale invoiced today is not cash until the customer pays, and a tax bill agreed months ago still lands on a specific day. Line those timings up across 13 weeks and you can see a shortfall weeks before it happens, which is exactly when you still have options. This guide shows you how to build one, phase the tax payments correctly, stress-test it against slow payers, and act early when it turns red.
Why 13 weeks, and not 4 or 52
Thirteen weeks is one quarter. That length is a deliberate compromise. Four weeks is too short to give you time to fix a problem: by the time you see the gap, the invoices you would chase are already overdue and the finance you might arrange has not cleared. A full year, by contrast, is mostly guesswork past the first few months, because you cannot know week 40’s sales or costs with any honesty.
A quarter is far enough ahead to act on. If week 9 shows a shortfall, you have roughly two months to chase debtors, delay non-urgent spending, arrange finance or agree a payment plan with HMRC. It is also short enough to stay accurate, because most of your near-term receipts and payments are already known: the invoices are raised, the direct debits are set, payroll is fixed.
The forecast should be rolling. Each week you drop the week just gone and add a new week 13 at the far end, so you always look a full quarter forward. Update it every week using real bank data, and compare what actually happened against what you predicted. That comparison is how the forecast gets sharper over time.
Profit and cash are not the same thing
Plenty of profitable firms go under, and the reason is almost always cash timing. Profit is an accounting measure: revenue earned minus costs incurred over a period, regardless of when the money actually moves. Cash is the balance in your bank account right now. The two can point in opposite directions for months.
Consider a services firm that wins a large project, invoices £40,000 on 30-day terms, and pays its contractors and staff to deliver it every week in the meantime. On paper it is highly profitable. In the bank it is bleeding, because wages go out weekly while the £40,000 arrives, if you are lucky, in five or six weeks. Growth makes this worse, not better: the faster you sell, the more cash you tie up in work you have delivered but not yet been paid for.
This is why the profit and loss account cannot tell you whether you can make payroll. Only a cash flow forecast can, because only it deals in the calendar. Understanding the gap also helps you judge which improvements genuinely pay off; our guide to the numbers that tell you automation has paid off uses the same cash-first logic.
How to build the forecast, line by line
Set up your weeks as columns and your line items as rows. Every week follows the same simple structure: opening balance, plus receipts, minus payments, equals closing balance. That closing balance becomes the next week’s opening balance. Get that carry-forward right and the rest is just careful listing.
Opening balance
Week 1’s opening balance is the actual cleared balance in your business current account today. Not the figure in your accounting software, which may include uncleared items; the real bank figure. Everything downstream depends on this being accurate, so reconcile first.
Receipts (money in)
List the cash you expect to collect each week, dated when you realistically expect it to land, not when you invoiced. Take your outstanding invoices (your debtors) and place each one in the week the customer actually tends to pay, which for a 30-day invoice from a habitually slow client might be week 7, not week 5. Add expected new sales, but be conservative about both the amount and the timing.
Separate reliable receipts from hopeful ones. A signed contract with a direct debit is near-certain. A verbal “we’ll probably renew” is not, and should sit in your slower scenario rather than your base case.
Payments (money out)
List every outflow on the week it leaves the account. The common lines for a small UK firm are:
- Payroll: net wages to staff on your pay date.
- PAYE and National Insurance: the tax and NI on those wages, paid to HMRC, usually by the 22nd of the following month.
- Pension contributions: your workplace pension payments.
- VAT: your quarterly return payment (covered below).
- Rent, rates and utilities: on their usual dates.
- Software subscriptions: the monthly and annual renewals that are easy to forget.
- Supplier and contractor invoices: dated when you will actually pay them.
- Loan and finance repayments: fixed direct debits.
- Corporation tax: the annual payment (covered below).
- Owner drawings or dividends: whatever you actually take.
If you are about to take on your first member of staff, the payroll, PAYE and pension lines all appear at once, so model them before you commit; our checklist for hiring your first employee in the UK sets out the full cost picture.
Closing balance
Closing balance equals opening balance plus receipts minus payments. Copy it into next week’s opening cell. Any week where the closing balance goes below zero, or below the buffer you need to sleep at night, is a red week. Those are the weeks the whole exercise is designed to expose.
Phasing VAT and corporation tax correctly
Tax payments are the outflows people most often get wrong, because they are large, irregular and easy to leave until they hurt. Phase them properly and they never surprise you.
VAT
Most VAT-registered businesses file quarterly. The payment is due one calendar month and seven days after the end of each VAT quarter, so a quarter ending 31 March is payable by 7 May. Put that lump sum in the correct week as a single payment. If you are on the flat rate scheme the sum differs, but the timing rule is the same. HMRC sets out the current deadlines on its VAT returns deadlines page. A good discipline is to move the VAT you collect into a separate pot as it comes in, so the payment week is funded rather than feared. Note that Making Tax Digital changes how many firms file; our SME guide to Making Tax Digital in 2026 covers who is affected.
Corporation tax
A small limited company pays corporation tax nine months and one day after the end of its accounting period. So a year ending 31 March is payable by 1 January. That is a single, often large, payment, and it needs its own row in the week it falls due. The rules are on the government’s pay your corporation tax page. Set money aside monthly against it, because trying to find the whole sum in one week is how otherwise healthy firms stumble.
A worked example: a small services firm
Here is a simplified five-week extract for a fictional consultancy with about £12,000 of monthly costs. A real forecast runs to 13 columns, but the mechanics are identical. All figures are illustrative.
| Line (£) | Week 1 | Week 2 | Week 3 | Week 4 |
|---|---|---|---|---|
| Opening balance | 8,000 | 10,500 | 6,700 | 4,900 |
| Receipts from debtors | 6,000 | 2,000 | 3,500 | 9,000 |
| Payroll and pension | 0 | 4,500 | 0 | 4,500 |
| Suppliers and subscriptions | 1,200 | 900 | 1,100 | 800 |
| Rent and utilities | 0 | 0 | 2,600 | 0 |
| PAYE / NI to HMRC | 0 | 0 | 0 | 1,600 |
| VAT payment | 2,300 | 0 | 0 | 0 |
| Closing balance | 10,500 | 6,700 | 4,900 | 6,400 |
Read across and you can see the shape of the business: payroll every fortnight, a VAT hit in week 1, rent in week 3, and a healthy week 4 receipt that pulls the balance back up. If that week 4 payment slipped, the firm would drop close to zero, which is exactly the risk the scenarios below are built to reveal.
Model three scenarios, not one
A single forecast gives false comfort. Build three versions of the same 13 weeks so you can see how much room you actually have.
- Base case: your honest best estimate, with debtor timings based on how customers actually pay, not your terms.
- Slow-paying customers: push every debtor receipt back by two to four weeks. This tests whether you survive a general slowdown in collections, which is the most common cause of a cash squeeze.
- Lost customer: remove your largest recurring customer’s receipts entirely from a chosen week onward. This shows your exposure to a single client and how many weeks you would have to react.
The gap between the base case and the two stressed versions is your margin of safety. If a modest delay in payments turns the account red in week 5, you are running with too little buffer and need to act before anything actually goes wrong.
What to do when the forecast goes red
A red week is a prompt, not a verdict. The earlier you spot it, the cheaper the fix. Work through these in order.
Chase your debtors first
The fastest cash is money you are already owed. Send statements, call the largest overdue accounts, and offer a card payment link to remove friction. Systematic chasing beats sporadic panic; automating reminders through your accounting tools helps, and our guide to invoice automation for UK firms covers how to get paid faster without nagging manually.
Tighten payment terms
For new work, shorten terms, take deposits, or bill in stages rather than at the end. A 50 per cent deposit turns a cash-hungry project into a self-funding one. This is a permanent fix, not a one-off.
Consider invoice finance
Invoice finance advances you a percentage of an unpaid invoice, typically most of its value, for a fee. UK banks and specialist lenders offer it, and it can bridge a genuine timing gap. It costs money and is not free cash, so use it for timing problems, not for propping up a business that is losing money.
Talk to HMRC early about Time to Pay
If a tax payment is the problem, contact HMRC before the due date, not after. A Time to Pay arrangement lets you spread a tax bill over an agreed period, and HMRC is far more receptive when you approach them ahead of the deadline. Details are on the government’s if you cannot pay your tax bill on time page. Silence is the worst option; an early call is often a straightforward one.
Automating the data feed
The chore that kills most forecasts is manual data entry. If updating it takes an hour every week, you will stop. The fix is to feed it from the software you already run.
If you use Xero, QuickBooks or Sage, your opening balance, outstanding invoices and recurring bills already live there. Dedicated forecasting apps connect to those accounts and pull the figures automatically, so your job shrinks to adjusting timings and reviewing scenarios. Modern accounting suites increasingly predict payment dates from a customer’s history; our roundup of the best AI accounting tools for UK small businesses covers which do this well. Keeping the source data clean matters more than the tool, and a tidy operations stack for a sub-£1m business makes that far easier.
Comparing your forecasting options
Three broad approaches exist. All can produce a good 13-week forecast; they differ in effort and automation. Prices are approximate and subject to change.
| Approach | Roughly | Strengths | Best for |
|---|---|---|---|
| Spreadsheet (Excel or Google Sheets) | Free or near-free | Total control, no limits, teaches you the mechanics | Owners who want to understand every line and have simple finances |
| Accounting software forecast (Xero, QuickBooks, Sage) | Included in most plans, from about £15 a month | Data already there, short-term projections built in | Firms wanting a quick view without a second subscription |
| Dedicated forecasting app (Float and similar) | From roughly £30 to £70+ a month | Automated feed, easy scenarios, clear visuals | Growing firms that update weekly and model several scenarios |
Start with a spreadsheet even if you later upgrade. Building one by hand teaches you how the numbers connect, which makes every automated tool afterwards more useful. A free UK-format template with the rows above is enough to begin this week.
The mistakes people actually make
Most forecasts fail for the same handful of reasons.
- Forecasting profit instead of cash. Dating receipts when you invoice rather than when you get paid quietly ruins the whole thing.
- Optimistic collection timings. Assuming customers pay on terms when they never have. Use their real history.
- Forgetting the lumpy payments. VAT, corporation tax, annual insurance and annual software renewals are the ones that ambush people.
- Building it once and abandoning it. A forecast is a weekly habit. A stale one is worse than none, because it gives false confidence.
- No buffer. Planning to hit zero means any small slip pushes you negative. Set a minimum balance and treat breaching it as the red line.
- Ignoring the stressed scenarios. The base case rarely happens exactly. The slow-payer and lost-customer versions are where the real decisions live.
Frequently asked questions
How often should I update my 13-week forecast?
Weekly, ideally on the same day each week. Replace the oldest week’s estimates with what actually happened, add a fresh week 13 at the end, and refresh your debtor timings. This rolling habit keeps it accurate and takes far less time than rebuilding it occasionally from scratch.
What if my income is irregular or project-based?
Irregular income is exactly when a 13-week forecast earns its keep. Enter only receipts you have real grounds to expect, weighted by how likely they are, and lean on the slow-payer scenario. Deposits and staged billing smooth the peaks and troughs, so build them into your standard terms.
Do I need accounting software to do this?
No. A spreadsheet and your online banking are enough to build a perfectly good forecast. Software mainly saves time by feeding the data automatically, which matters most once weekly updates start to feel like a chore.
How is this different from a budget?
A budget is an annual plan for revenue and costs, usually measured monthly and focused on profit. A 13-week cash flow forecast is a short-term, weekly view of actual bank movements. You need both: the budget sets direction, the forecast keeps you solvent along the way.
What buffer should I keep in the account?
There is no universal figure, but many owners aim for enough to cover several weeks of fixed costs such as payroll and rent. Set the number you personally need to feel secure, mark it on the forecast, and treat any week that dips below it as a signal to act.
What to do next
- Build the skeleton this week. Open a spreadsheet, create 13 week columns, add the rows from this guide, and enter today’s real bank balance as week 1’s opening figure.
- Populate it with real timings. Add your outstanding invoices dated when customers actually pay, then every known payment including payroll, PAYE, VAT and any corporation tax due in the quarter.
- Run the two stress scenarios. Copy the sheet twice: push all receipts back a few weeks in one, and remove your biggest customer in the other. Note the first red week in each.
- Set a weekly review and act on red early. Book a 20-minute slot each week to update it, and the moment a red week appears, start chasing debtors or, if tax is the issue, contact HMRC about Time to Pay before the deadline.
