Key takeaways
What this article covers, in order:
- Why 13 weeks and not a year?
- What goes into the forecast
- A worked example
- What Callum does about weeks 11 and 12
- Building your own, step by step
- Keeping it honest: the weekly 15 minutes
A 13-week cash flow forecast is a week-by-week list of the money you expect to come in and go out over the next quarter, ending with a projected bank balance for each week. It's short enough to be accurate and long enough to spot a crunch before it arrives. Here's a 13 week cash flow forecast template you can build in an afternoon and keep up to date in fifteen minutes a week.
Why 13 weeks and not a year?
Callum runs a landscaping business near Cardiff with two staff and a van that's older than one of them. He's tried annual forecasts. By the third month they were fiction. A customer paid late, a mower died, a big job slipped into the following month. The spreadsheet looked lovely and told him nothing.
Thirteen weeks is a quarter. You know most of your customers, most of your bills and most of your fixed costs that far out. You can see the quarterly rent, the insurance renewal and the quiet weeks around Christmas. And because it's weekly, a single late payment shows up as a dip you can do something about, rather than vanishing into a monthly average.
What goes into the forecast
Keep it to cash. If money doesn't move in or out of the bank that week, it doesn't go in.
Cash in
- Customer invoices you expect to be paid, in the week you realistically expect the money (not the due date if that customer is always late)
- Card or till takings if you have them
- Any one-offs: a loan drawdown, a grant, the owner putting money in, selling an old vehicle
Cash out
- Wages and anything paid alongside them
- Rent, rates and utilities
- Supplier bills, in the week you plan to pay them
- Loan and finance repayments
- Insurance, subscriptions, software
- Tax payments your accountant has told you about
- Money the owner takes out
- Planned equipment purchases
You'll notice there's no depreciation and no "income earned but not yet invoiced". This is a cash forecast, not a P&L.
A worked example
Here's Callum's forecast starting Monday 12 Oct 2026, with £8,000 in the bank. He's decided he never wants to drop below £4,000, because that's roughly two weeks of wages and fuel.
| Week | Week starting | Cash in | Cash out | Net | Closing balance |
|---|---|---|---|---|---|
| 1 | 12 Oct 2026 | £6,200 | £5,400 | £800 | £8,800 |
| 2 | 19 Oct 2026 | £4,100 | £3,900 | £200 | £9,000 |
| 3 | 26 Oct 2026 | £3,800 | £6,100 | −£2,300 | £6,700 |
| 4 | 2 Nov 2026 | £7,400 | £4,800 | £2,600 | £9,300 |
| 5 | 9 Nov 2026 | £2,900 | £3,500 | −£600 | £8,700 |
| 6 | 16 Nov 2026 | £4,500 | £7,200 | −£2,700 | £6,000 |
| 7 | 23 Nov 2026 | £3,600 | £3,300 | £300 | £6,300 |
| 8 | 30 Nov 2026 | £6,800 | £4,600 | £2,200 | £8,500 |
| 9 | 7 Dec 2026 | £3,200 | £3,800 | −£600 | £7,900 |
| 10 | 14 Dec 2026 | £2,400 | £5,200 | −£2,800 | £5,100 |
| 11 | 21 Dec 2026 | £1,100 | £2,600 | −£1,500 | £3,600 |
| 12 | 28 Dec 2026 | £600 | £1,400 | −£800 | £2,800 |
| 13 | 4 Jan 2027 | £5,900 | £4,100 | £1,800 | £4,600 |
| Total | £52,500 | £55,900 | −£3,400 |
Opening £8,000, plus £52,500 in, minus £55,900 out, gives £4,600 at the end of week 13. The columns tie.
A few of the lumps explained:
- Week 3 is the quarterly rent on his yard plus a fuel card payment.
- Week 6 has the van insurance renewal on top of wages.
- Week 10 is wages plus a Christmas bonus he'd half-promised.
- Weeks 11 and 12 are the problem. Customers wind down, nobody pays invoices between Christmas and New Year, and the balance falls to £2,800, well under his £4,000 floor.
What Callum does about weeks 11 and 12
The point of a forecast isn't to be right. It's to give you options while there's still time to use them. Seeing the dip in the second week of Oct 2026 gives Callum ten weeks to act. His choices:
- Chase earlier. Three customers usually pay in the first week of Jan 2027. He sends invoices for the Nov 2026 work promptly on 30 Nov 2026 with 14-day terms instead of 30, and politely reminds them before the holidays.
- Move the bonus. Paying the bonus in week 13 instead of week 10 lifts weeks 10 to 12 by the same amount.
- Ask a supplier for a few extra days. His turf supplier is fine with mid-Jan 2027 for the Dec 2026 order if he asks now, not on Christmas Eve.
- Line up a fallback. If all else fails, he knows he might need his overdraft for a fortnight, and he can talk to the bank calmly in Oct 2026 rather than in a panic in late Dec 2026.
He doesn't need all four. Two of them fix it.
Building your own, step by step
- Start from a reconciled bank balance. If the opening figure is wrong, every week after it is wrong. Reconcile first, then forecast.
- List the fixed outgoings. Wages, rent, loans, insurance, subscriptions. These are the easy, reliable rows.
- List known invoices. Go through aged debtors and drop each invoice into the week you honestly expect payment. Be a bit pessimistic about known late payers.
- Estimate new sales. Use the last few months as a guide. Leave some slack; don't forecast your best month as normal.
- List supplier bills. From aged creditors, place each one in the week you intend to pay it.
- Add the one-offs. Equipment, owner drawings, any tax payments your accountant has flagged.
- Set your floor. Pick a minimum balance you're comfortable with and highlight any week that drops below it.
A spreadsheet is fine for this. Thirteen columns, a row for each type of cash in and out, a net line and a running balance.
Keeping it honest: the weekly 15 minutes
A forecast that isn't updated is worse than none, because you'll trust it. Every Monday:
- [ ] Reconcile last week's bank transactions
- [ ] Replace last week's forecast numbers with what actually happened
- [ ] Note any big differences and why (late payer, unexpected repair)
- [ ] Roll forward: drop the finished week, add a new week 13 at the end
- [ ] Re-check the lowest balance and the week it falls in
- [ ] Decide one action if anything is under your floor
The "why" column is where you learn. After a couple of months you'll know which customers pay on the 45th day no matter what your terms say, and you can forecast them that way.
Common mistakes
Forecasting invoices on their due date. Use the date the customer actually tends to pay. Your aged debtors history tells you.
Forgetting annual and quarterly bills. Insurance, software renewals, accountant's fees, the MOT. Put a reminder in for anything that doesn't come monthly.
Mixing up profit and cash. A big job won this month doesn't help the bank until it's paid.
Not including what you take out. If you pay yourself, it's cash out. Leave it off and the forecast looks rosier than reality.
Starting from an unreconciled balance. It's tempting to grab the number from your banking app. Fine, but make sure your books agree with it, or the forecast for the business and the books will drift apart.
How HelloBooks helps
HelloBooks doesn't build the 13-week forecast for you, but it supplies the numbers you need for it. On every plan, including Free, you get aged debtors and aged creditors reports (the cash in and cash out you already know about) and a Cash Flow report showing what actually happened.
Your bank connects through Open Banking for most UK banks and cards, or you can import a CSV statement. The reconcile screen lines statement lines up against your ledger with AI match suggestions, so your opening balance each Monday is one you can trust. Free includes one live bank feed; Pro (£14.99/month) has unlimited bank connections, recurring invoices and bills with approvals, which makes the known rows of the forecast easier to fill in. More on the cash flow management page and on pricing.
FAQs
Why is it called a 13-week forecast?
Thirteen weeks is one quarter of a year. It's long enough to see quarterly bills and seasonal dips, and short enough that your figures are mostly based on things you actually know rather than guesses.
Should I forecast weekly or monthly?
Weekly, for a 13-week view. Monthly totals hide the problem: you can be fine at the end of the month and still short on the Friday wages go out. Weekly shows the low point.
How big should my cash buffer be?
There's no single rule. Many small businesses aim for enough to cover two to four weeks of unavoidable costs like wages and rent. Pick a number that would let you sleep if a big customer paid a fortnight late.
What if my forecast is wrong every week?
That's normal at the start. Track the differences and the reasons. Most errors come from a handful of late payers or forgotten bills, and once you adjust for them the forecast gets noticeably closer within a couple of months.
Do I need accounting software to do this?
No, a spreadsheet works. What you do need is a reconciled bank balance and up-to-date lists of who owes you and who you owe. That's where software saves the time.
Should my accountant check it?
If you're using the forecast to apply for a loan or to decide whether to hire, it's worth a second pair of eyes. For week-to-week planning, you're the best person to keep it.
Fifteen minutes on a Monday beats a sleepless night in late Dec 2026.
Start free, no card needed. Try HelloBooks Free
