Cash flow for UK businesses: a complete guide

Cash flow is the money actually moving in and out of your bank account, as opposed to the profit your accounts report. Working capital is what you hold to bridge the gap between paying your costs and getting paid: current assets minus current liabilities, on the British Business Bank's definition, and that definition includes your cash. Most of what you can influence is on the receivables side.

In this article

The short version

• Working capital is current assets minus current liabilities on the British Business Bank's definition, and that version includes cash, so say which definition you used before you compare your figure to anyone else's.
• No UK government body publishes debtor days or DSO for small businesses, by sector or by size. The only official payment-timing statistic runs the other way, and it is unaudited.
• The widely quoted "average time to pay: 32 days" for large businesses is a median on the department's own methodology report, from self-reported data with "no independent verification or auditing".
• One in seven (14%) trading UK businesses had no cash reserves in late December 2025 (ONS Business Insights and Conditions Survey, wave 147, 39,925 businesses).
• The stock of SME overdrafts stood at £7bn in December 2025, down 54% in nominal terms since 2012 and the lowest on record (British Business Bank, Small Business Finance Markets 2025/26).
• The 13-week rolling forecast is a vendor convention. The only horizons any UK authority publishes are the FRC's 12 months for going concern and ICAEW's 18 months for working capital statements in capital markets transactions.

Cash flow, working capital and profit

Your profit and loss account tells you whether the work was worth doing. Your bank balance tells you whether you can pay for it this week. A business can be profitable on paper and still have nothing available for a supplier, because an invoice counts as revenue the day you raise it and as cash on the day it clears.

Working capital is the buffer between those two positions. The British Business Bank defines it as "the money your business needs to be able to operate from day to day", with the arithmetic given as "Working capital = current assets - current liabilities" and the ratio as "current assets / (divided by) current liabilities", on a page that carries no date.

One thing to pin down before you quote your own number. That same page uses "working capital" and "net working capital" interchangeably, where much of the finance literature reserves net working capital for current assets minus current liabilities and uses operating or trade working capital for inventory plus receivables minus payables, leaving out cash and debt. The British Business Bank's version includes your cash, so the two calculations differ by your entire cash balance. A working capital figure from your accounts means very little until you say which one produced it.

For the statutory cash flow statement, IAS 7 defines cash as "cash on hand and demand deposits" and cash equivalents as "short-term, highly liquid investments that are readily convertible to known amounts of cash" carrying "an insignificant risk of changes in value". Money in a 90-day notice account isn't cash for that purpose, and it isn't cash for your Thursday payment run either.

The cash conversion cycle, and whose formula it is

The cycle is the interval between money leaving you and money coming back. You pay your staff and suppliers, you deliver, you invoice, and at some later point your customer pays. The British Business Bank describes it as "the time it takes to get paid after you've incurred costs in delivering a product or service", and it publishes no formula for it.

Nor does anybody else in an official position. No UK primary source we could fetch publishes a formula for the cash conversion cycle, for debtor days or DSO, or for creditor days. We looked at the British Business Bank, the Bank of England, ONS, the business department, the fetchable ICAEW pages and the Small Business Commissioner. The formulas everybody quotes are textbook conventions. Nothing about that makes them unreliable, and it does make publishing your own workings worthwhile.

So when you calculate debtor days, state the arithmetic you used. On the conventional calculation it's trade debtors ÷ annual credit sales × 365. A countback method, where you work backwards through recent sales until the debtor balance is used up, gives a different answer on the same ledger, and no UK authority prescribes either. If you're tracking the figure month to month, your own consistency matters more than which method you picked.

Two siblings go further into this: the credit control metrics worth tracking and how to reduce debtor days in a small business.

Where receivables come into it

The cycle has three legs: what you pay out, how long delivery takes, and how long the money takes to come back. On the first two you're negotiating with suppliers and running your own operation. On the third, somebody outside your business decides the timing.

That asymmetry is why receivables get the attention. You control when the invoice goes out, what it says, how easy it is to pay, and what happens after the due date. You don't control the date your customer's finance team runs its payments, though you can find out when that is and get the invoice in front of it. Reading your own ledger properly comes first, and how to read an aged debtor report covers the report most accounting software gives you.

What the official statistics say, and what they leave out

Start with what doesn't exist, because it will save you time. No UK official statistic for debtor days or DSO exists, for any sector or any size band. Every "average debtor days for small businesses" figure in circulation comes from accounting-software ledgers or from a survey. Both can be useful; neither is official.

What does exist runs in the opposite direction. Large businesses have to report how long they take to pay their own suppliers, and those reports are published as Official Statistics. The 2025 edition, published on 14 July 2026, gives an average (median) time to pay of 32 days, with 15% of invoices paid late by number and 14% by value. Manufacturing was slowest at 45 days with the highest late rate at 21%, and financial and insurance fastest at 21 days with 13% late. London was shortest at 27 days and the East and West Midlands both 38 days.

Three qualifications come with that 32 days, all of them from the department itself.

1. The 32 days is a median described as an average. The commentary calls it the average time to pay, while the methodology and background quality report states that "Output statistics are based on median values". So the headline is a median of self-reported medians.
2. The data is unaudited. Verbatim: "There is no independent verification or auditing of the data. This means that some reports may be submitted incorrectly or contain mistakes." And "Not all businesses required to report do so", so the output "only describe the performance of businesses that report".
3. It covers large payers only. A business reports if it exceeds two of three thresholds on its last two balance sheet dates: "£54 million annual turnover", "£27 million balance sheet total", "250 employees". The 2025 scope changes also reduced the reporting population by 6.5%, so year-on-year comparisons aren't clean.

The practical value here isn't the national average. Every individual report is searchable, so you can look up your own biggest customer before you agree terms, on the government's Check Payment Practices service.

Two figures you'll meet elsewhere are worth labelling. Xero publishes that "The average length of time small businesses waited to be paid, after issuing an invoice, was 29.1 days" and that they "were paid, on average, 8.3 days late" in the June quarter 2026: that's Xero's own customer data and the page carries no sample size. Good Business Pays figures appear on the Small Business Commissioner's news pages, and a campaigning group's research carried there isn't the Commissioner's own.

How much cash do UK businesses hold?

The one regular official statistic about UK businesses' cash position comes from the ONS Business Insights and Conditions Survey. In wave 147, covering 15 to 28 December 2025 and published on 8 January 2026: "Approximately one in seven (14%) trading businesses reported they had no cash reserves in late December 2025". The sample was 39,925 businesses with a 25.9% response rate, and ONS describes the survey as voluntary and fortnightly, so treat it as a large voluntary sample and not a census. That figure describes the consequence more directly than any late payment statistic does: one business in seven has no buffer at all, so a single invoice arriving three weeks late lands on a bank balance with nothing behind it.

What you can influence and what you can't

Most cash flow advice mixes the two together, so here they are separately.

The right-hand column isn't a list of things to ignore. You can look up a large customer's published payment performance, ask a new client when their payment run falls, and price the risk of a slow payer into the job. What you can't do is move any of it by chasing harder.

The funding around the gap

The buffer you may remember has largely been withdrawn. The stock of outstanding SME overdrafts held by banks "fell by 54% in nominal terms" between 2012 and 2025, and stood at "£7bn in nominal terms" at December 2025, "down 13% from 2024, the third annual fall, and the lowest value on record" (British Business Bank, Small Business Finance Markets 2025/26, using data to 12 February 2026). Half of all smaller businesses use external finance and 50% of those seeking it do so to support working capital, while asset finance flow reached "£24.4bn in nominal terms" in 2025, the highest on record. The money hasn't gone, then. It has moved from a facility you could dip into towards finance secured against something.

Invoice finance is aimed squarely at your receivables, and the two authorities on it disagree about the headline number. The British Business Bank says a client "will then have access to up to 80 or 90 per cent of the value of those invoices virtually immediately". ICAEW's best-practice guideline 65, published in 2017, gives advance rates on invoices as "70-90%". Both are primary-grade sources published nine years apart, so treat 70% to 90% as the range and name both sources when the number matters to you.

The cost structure is published in words and not in percentages. You pay "a service fee (a percentage of the value of each invoice), and a discount charge (like interest) based on the amount of funds it utilises". Neither authority publishes a figure for either component, so any percentage you're quoted comes from the provider quoting it. The two product shapes differ in who talks to your customers. With factoring the provider "will manage your sales ledger and be involved in collecting payment for your invoices direct from your customers", and "Your customers are likely to know" about it. Invoice discounting is "a finance-only product" and "many invoice discounting facilities are undisclosed". ICAEW's 2017 guideline adds that for many SMEs "the costs of an invoice discounting facility are likely to be broadly competitive with overdrafts and conventional loans."

One number nobody can give you is the price of an overdraft. The Bank of England publishes effective interest rates for SME loans, on both the stock outstanding and new lending, but overdrafts appear only under the broader private non-financial corporations category. Bank Rate was 3.75% on 19 August 2026, so if you're financing unpaid invoices on an overdraft priced at Bank Rate plus a margin, you're paying that margin plus 3.75%. Any article telling you what late payment costs you in interest has assumed a rate nobody publishes.

Forecasting, and the only horizons anyone has published

Three horizons circulate, and they have very different standing.

The 13-week rolling forecast is the most repeated, and every source for it is a software vendor or a vendor's blog. No UK regulator, professional body or government source publishes it, prescribes it, or explains why thirteen weeks. It may still suit you, and there's nothing behind it beyond convention.

The FRC does specify a horizon, for going concern. Its February 2025 guidance says "the minimum period for the going concern assessment is 12 months from the date the financial statements are authorised for issue", and that "A longer assessment period could be more appropriate" where events or conditions are identified beyond that minimum. A forecast's job, in its words, is to "indicate whether there is an adequate matching (of both the timing and the amount) of projected cash inflows with projected cash outflows", and the guidance is non-mandatory.

ICAEW goes further out in one context. In TECH 04/20CFF, on working capital statements in capital markets transactions, directors are asked to prepare projections covering "at least the next 18 months" and to address "foreseeable working capital difficulties beyond the 12-month period". That's capital markets guidance and not general small business practice, and in the one place a UK body does prescribe a horizon it asks for six months more than the period being formally opined on.

One line in that guidance is about the forecaster instead of the forecast. The FRC: "Directors should be aware of their own potential bias when making judgements and assumptions in forecasts." Nobody selling you a forecasting template opens with that. The same guidance gives you three ways to test your assumptions: sensitivity analysis, "measuring the impact on forecasts of changing individual assumptions"; stress testing, "identifying adverse scenarios and understanding their effects"; and reverse stress testing, a test that "typically starts from a situation that causes a business to fail".

Building the receipts side of a forecast when your customers pay unpredictably is a job of its own, and how to forecast cash flow when clients pay late does it properly.

What late payment costs, with the basis of each figure

Four government figures get quoted constantly, and each carries more caveats than the headline suggests. All four come from Late payments research: impact on the UK economy, published in July 2025 by the Department for Business and Trade and the Office of the Small Business Commissioner, written by London Economics on a YouGov and IFF Research survey of 1,455 businesses, fieldwork 15 January to 7 February 2025.

• Staff time: "22% of surveyed businesses said they spent staff time chasing late payments, on average 86hrs per business affected by late payments per year", totalling 133 million hours across the economy. That is 86 hours per business affected, among the 22% who reported spending time on it, and most quotations drop both qualifiers.
• Money outstanding: an estimated £26bn owed at any one time, on a 90% confidence interval of £18bn to £36bn, averaging £17,000 per affected business.
• Economy-wide cost: almost £11bn a year, a modelled aggregate of six cost channels, five of them survey self-report. The authors say it "could be conservative" in places and could overstate in others, in the same report.
• Closures: the department and the Office of the Small Business Commissioner estimate about 14,000 additional business closures a year, or 38 a day, attributable to late payment. That comes from statistical matching on 2015 to 2017 survey data linked to the business register, and the authors record that "none of the estimates above are statistically significant" when broken down by firm size, with a 90% confidence interval on the cost running from £437m to £5,979m.

The older claim that 50,000 businesses fail each year because of late payment can't be traced to any current primary source, and the official 14,000 supersedes it.

What you can act on is covered in the late payment set: UK late payment rules, late payment interest and compensation and recovering overdue debt, the steps in order.

The rest of this category

Five articles go deeper than this one, each on a question this guide only opens for you.

• How to forecast cash flow when clients pay late: building a receipts forecast from the shape of your lateness instead of an average.
• How to reduce debtor days in a small business: the levers on the receivables leg.
• Lock-up in an accountancy practice: the practice version of the same measurement problem.
• The case for zero WIP and zero debtors in a practice: billing and collecting so the cycle barely opens.
• How long does it take to get overdue invoices under control?: what changes over the first months of collecting properly.

Your influence is concentrated in the receivables leg, and that leg depends on the invoice going out promptly, the payment being easy, and the follow-up happening whether or not you remembered it. Adfin does that side of it for you: invoicing, collection by card, open banking, bank transfer or direct debit, and chasing, in one place. To see how that would work on your own ledger, book a demo.

Common questions

What's the difference between cash flow and profit? Profit measures whether the work was worth doing over a period; cash flow measures what actually moved through your bank account and when. An invoice counts as revenue when you raise it and as cash when it clears, so a profitable business can still be unable to pay a supplier.

What is working capital, and does it include cash? Working capital is current assets minus current liabilities on the British Business Bank's definition, and that definition does include cash. Much of the finance literature uses operating or trade working capital for inventory plus receivables minus payables, leaving out cash and debt, so the two calculations differ by your whole cash balance. Say which one you used.

What is the average debtor days for a UK small business? No UK government body publishes it, for any sector or size band. Every figure you'll see comes from accounting-software ledgers or a survey, and none is official. Xero publishes 29.1 days waited and 8.3 days late for the June quarter 2026, from its own customer data and with no sample size on the page.

How long do large UK businesses take to pay? The 2025 official statistics, published on 14 July 2026, give an average (median) time to pay of 32 days, with 15% of invoices paid late by number. The headline is a median described as an average, it's self-reported by the payers, and the department states there is "no independent verification or auditing of the data".

How far ahead does a cash flow forecast need to look? No UK authority prescribes a horizon for a management forecast. The FRC requires a minimum of 12 months for a going concern assessment, and ICAEW asks for at least 18 months for working capital statements in capital markets transactions. The 13-week rolling forecast is a vendor convention with no professional body behind it.

Is invoice finance worth it for cash flow? It converts an unpaid invoice into cash now at a cost, so it helps with timing and not with a customer who never pays. Advance rates run from 70% to 90% of invoice value: the British Business Bank says up to 80 or 90 per cent, ICAEW's 2017 guideline says 70 to 90 per cent. Neither publishes the service fee or discount charge, so compare quotes on total cost.

Sources

• IFRS Foundation — IAS 7 Statement of Cash Flows (accurate as of August 2026)
• FRC — guidance on the going concern basis of accounting (accurate as of August 2026)
• Check Payment Practices service (accurate as of August 2026)
• GOV.UK — large businesses payment practices and performance statistics 2025 (accurate as of August 2026)
• ONS — Business insights and impact on the UK economy, 8 January 2026 (accurate as of August 2026)
• British Business Bank — why working capital is important to your business (accurate as of August 2026)
• British Business Bank — Small Business Finance Markets 2025/26 (accurate as of August 2026)

This article is information, not advice. The working capital definition and formulas, the invoice finance descriptions and advance range, and the SME overdraft and asset finance figures come from the British Business Bank, the last from Small Business Finance Markets 2025/26 using data to 12 February 2026. Cash and cash equivalents are defined from IAS 7 as published by the IFRS Foundation. The 32 days, the late payment proportions, the sector and regional figures, the reporting thresholds and the audit limitation come from the Large businesses' payment practices and performance statistics 2025, published 14 July 2026, with the median wording from the same series' methodology and background quality report. The 14% with no cash reserves is ONS Business Insights and Conditions Survey wave 147, 39,925 businesses, 25.9% response rate, published 8 January 2026. The advance rate range of 70 to 90 per cent, the service and discounting charge structure and the comparison with overdrafts come from ICAEW best-practice guideline 65, published 2017. Forecast horizons and the bias quotation come from the FRC's February 2025 going concern guidance, non-mandatory, and ICAEW TECH 04/20CFF, capital markets guidance. The late payment figures come from Late payments research: impact on the UK economy, July 2025, by the Department for Business and Trade and the Office of the Small Business Commissioner, written by London Economics on a survey of 1,455 businesses. Bank Rate was 3.75% as of 19 August 2026 and no UK SME overdraft rate is published. Last updated August 2026.