A profitable business runs short of cash because profit and cash are measured at different moments. The gap sits in specific places: invoices raised but not collected, stock or work in progress bought ahead of the sale, and outflows that never reach the profit and loss at all. Each one leaves a different signature, so each one is diagnosable.

Why A Profitable Month Can Still End Short

Accrual accounting recognises revenue when you have earned it and costs when you have incurred them. Cash moves on a separate schedule. That is not a flaw in the method; it is what the method is for. But it does mean your profit and loss answers a question you are not asking when the bank balance surprises you.

The difference between the two does not vanish. It accumulates in a handful of balance-sheet accounts, and it helps to think of them as waiting rooms between recognition and settlement. Receivables hold revenue you have recognised and cash the customer still owes. Inventory and work in progress hold cash already spent on something not yet sold. Payables and accruals hold costs you have recognised and not yet paid, flattering the bank balance until the payment run lands. Prepayments hold cash that left in one month for a cost spread across twelve.

The direction is not fixed, which is worth saying because most explanations assume it runs one way. A business billing on 60-day terms recognises profit in month one and sees the cash in month three. A business paid annually in advance sees the cash in January and recognises revenue across the year. Same mechanism, opposite shape.

Growth Consumes Cash Even When Nothing Goes Wrong

The most common version of this problem is not a failure at all. It is growth funded out of working capital, with every operational metric holding steady.

Take a business at £6m revenue with £3.9m cost of sales, collecting in 45 days, holding 40 days of stock or work in progress, and paying suppliers in 30 days. Grow revenue 20% to £7.2m with no change to any of those days:

That is arithmetic rather than a benchmark — apply your own days and volumes and the shape holds. Twenty per cent growth absorbs roughly £169,000 while the business does everything right. Debtor days did not move. Nobody paid late. The money went into funding a larger operating cycle.

This is why a stable debtor-days figure proves less than finance teams expect: stable days on a bigger base still means more cash tied up. So when you decompose the movement, always separate the volume effect (more sales, same days) from the efficiency effect (days getting worse). They point at different management problems, and the efficiency effect is usually the larger. In the example above, if receivables days also slipped from 45 to 55, that ten-day slip would absorb a further £197,000, more than the entire cost of the growth.

The Outflows That Never Touch Your Profit And Loss

The second common explanation is not on the profit and loss at all, which is exactly why it surprises people who have just read the profit and loss.

Capital expenditure leaves the bank in full on purchase day and returns as depreciation over years. Loan principal repayments reduce a liability; only the interest is an expense. Dividends and drawings are distributions, not costs. Corporation tax is accrued across the year and settled on a statutory date unrelated to when the cash was earned — in the UK, normally nine months and one day after the accounting period end for companies below the instalments threshold. And indirect tax is money you collected on someone else's behalf, held, and handed over in one movement.

That last one needs its own treatment, because our four markets handle it differently and the differences are not cosmetic.

UK readers often ask about the VAT Cash Accounting Scheme, which moves the tax point to the payment date and removes the problem of funding VAT on unpaid invoices. The ceiling matters: you can join with taxable turnover up to £1.35m and must leave once it exceeds £1.6m. Most businesses in the £2.5–10m band are already above it and have to manage the mismatch through forecasting instead. Leaving the scheme also triggers a one-off catch-up on outstanding invoices, usually in the quarter a growing business can least afford it.

How To Diagnose Your Own Cash Gap In Seven Checks

Every article on this subject lists causes. The useful question is which one is yours, and that comes from running an ordered sequence rather than debating theories.

The order is deliberately cheapest and most common first, so data problems and timing effects are ruled out before anyone commissions a customer-profitability review. It reflects cost and isolating power, not a claim about which cause is most frequent — no published evidence supports a universal ranking for businesses of this size.

Step 1: Reconcile every bank account before explaining anything. Confirm the ledger matches the bank for each account and entity, and check transfers in transit, duplicate imports, unallocated receipts and uncleared payments. If the gap disappears here it was a data problem, and a surprising number are.

Step 2: Bridge the cash movement to profit. Work from last period's closing cash through operating profit, non-cash items, the movements in receivables, inventory and payables, then cash tax, capital expenditure, loan principal and drawings. Every material movement must land in one of those buckets; what is left over is what you have not yet explained.

Step 3: Split each working-capital movement into volume and days. For receivables, inventory and payables separately, work out how much of the change came from higher volume at unchanged days and how much from the days deteriorating. A £150k receivables increase from growth is a financing question; the same increase from slower payment is a collections question.

Step 4: Measure how customers actually pay, not how they agreed to pay. Use paid invoices rather than the ageing report, and calculate days beyond terms per customer from invoice date, due date and cleared date. Read the median and the slow tail, not the average — a few large late payers move your cash more than the typical customer. Published figures on how far terms and behaviour diverge disagree with each other, so measure your own.

Step 5: Schedule the committed outflows that never appear in profit. List the next thirteen weeks of VAT, GST or sales tax, corporation tax, payroll, loan principal, capital expenditure, drawings, annual renewals and intercompany settlements by the date each leaves the bank. Concentration is the risk — several landing in one week is a gap even when the quarter looks comfortable.

Step 6: Classify every forecast variance rather than counting them. Mark each material difference as a timing variance, a permanent variance, an omission, a double count, a cut-off error or a data-latency difference. This is the cleanest boundary in the exercise: timing variances are a forecasting and collections problem, permanent variances are a business problem, and they do not respond to the same fixes.

Step 7: Only now test whether it is margin rather than timing. If steps one to six explain the movement, you have a timing issue. If cash is still eroding, look at gross margin by product, customer and project, at overhead run rate, and at whether growth is outrunning available funding. It sits last because it is the most expensive check to run, not the least likely to matter.

Run it in that order and most gaps resolve in the first three steps. Run it monthly and it stops being an investigation and becomes a routine — the rhythm that makes an early warning system for cash flow problems work.

Why Your Monthly Forecast Keeps Missing

If the gap keeps recurring, the forecast that failed to warn you deserves examining too. The failures are consistent.

It forecasts revenue rather than collections. A revenue forecast says when sales will be recognised, not when money will clear, and growing businesses commit against cash that has not arrived.

It uses accounting dates rather than expected payment dates. Invoice date, due date and expected payment date are three different things, and only the third belongs in a cash forecast.

It leaves out the outflows above. Capital expenditure, loan principal, tax and drawings are absent from anything derived from a profit and loss, and they are the most predictable surprises you will have.

It is never re-forecast, so it cannot learn. Without comparison against actuals there is no variance data, and your timing assumptions never improve.

Different systems carry different clocks. The bank has posted a payment, the accounting feed has not imported it, and a planning tool is showing last night's sync. All three can be right at once.

One honest note on accuracy. Published benchmarks for how accurate a 4, 13 or 26-week forecast should be do not survive checking: the circulated figures trace back to other content rather than to studies with disclosed samples and methods, and we excluded every one we found. Measure your own error instead — variance in closing cash at one, four, eight and thirteen weeks ahead, with inflow and outflow variance tracked separately.

The same applies to re-forecasting frequency. The case for a weekly rhythm is that it generates variances you can classify and learn from, not that frequency mechanically improves accuracy; the evidence for that is thinner than our category generally admits. Weekly works because it creates the feedback loop, which is the argument for a rolling 13-week cash flow forecast.

When The Change Is Sudden

A cash position that moves overnight is a different question and needs a different tool. Work from the bank outwards, not from the forecast inwards.

Start in the bank's own portal and identify the transactions behind the movement, separating pending from cleared. Reconcile the ledger to the bank to establish whether the movement is real or an artefact. Check one-off outflows first — a tax payment, loan repayment, capital purchase, drawing, or a supplier run pulled forward. Then check which expected receipts failed to arrive, and whether cash arrived but sits unallocated. Only after that do the ratios tell you anything.

What usually goes unsaid is data latency, and it matters because each layer carries a different timestamp. The bank's portal is closest to the truth. Your accounting platform sits behind it: Xero's own guidance is that bank feeds refresh overnight, and feeds using multi-factor authentication must be refreshed by hand, while QuickBooks Online attempts an update roughly every 24 hours and does not download pending transactions. Anything that syncs from your accounting platform inherits that lag and adds its own.

So a forecasting layer is the wrong first instrument for an overnight movement. It works from data that has cleared your accounting platform, which is the right basis for planning and the wrong basis for finding out what happened an hour ago. Record the as-of time for each layer alongside the number, and never use a downstream dashboard to disprove a bank movement that has not yet reached the ledger.

What Actually Closes A Cash Gap

Match the remedy to the diagnosis, cheapest first.

Process changes cost nothing but discipline. Invoice on the day the obligation is met rather than in a month-end batch. Drive collections from your own days-beyond-terms data rather than ageing buckets, so the customer who always pays 20 days late is chased before the invoice is technically overdue. Take deposits or bill in stages on larger work. Align payment runs to due dates rather than paying early out of habit. Ring-fence collected tax so it is never counted as available cash.

Financing closes a timing gap faster and costs money. An overdraft or revolving facility suits a stable, capped, intermittent gap and charges on the drawn balance. Invoice finance suits a growing gap shaped by slow-paying customers, and typically combines a service charge on the ledger with a discount charge over base rate on funds drawn. Which is cheaper depends on the shape of your gap, and the published cost comparisons come from lenders and brokers and disagree with each other — get quotes against your own numbers rather than a range in an article.

Sometimes it is not a timing problem at all. When the seven checks come up empty, look for margin erosion, for over-trading where each new unit of revenue needs more working capital than the business retains in profit, or for a customer or product line that is unprofitable once fully costed. Financing a structural problem makes it larger and more expensive. No amount of forecasting accuracy makes an unprofitable contract profitable; it only tells you sooner.

How Float Fits

Float is a cash forecasting layer on top of Xero or QuickBooks Online, built for finance teams running this diagnosis on a weekly rhythm.

It syncs with your accounting platform every 24 hours, with a manual sync when you need one, and builds the forecast from data that has cleared that platform — reconciled bank balances, invoices, bills and transactions. It does not connect to your bank directly, so it is not the tool for investigating what posted this morning. What it does give you is every invoice and bill mapped to its expected payment date rather than its accounting date, which is the single change that fixes most of the forecasting failures above.

From there you get a rolling 13-week view for operational decisions and a monthly view extending up to three years for board and budget conversations, scenario planning to test a decision before committing to it, and consolidation across multiple entities and accounts into one position. Permissions are role-based — Admin, Editor and Viewer — and two-factor authentication is mandatory for Xero users. Sage Intacct is on the waitlist rather than live.

If your team is building this capability from scratch, our guides on implementing cash flow forecasting in a finance team and improving cash flow visibility cover the operating model around the tool.

Frequently Asked Questions

Why is my business profitable but has no cash?

Profit and cash are recognised at different moments, and the difference accumulates elsewhere. The usual places are invoices raised but not collected, stock or work in progress funded ahead of the sale, and outflows such as capital expenditure, loan principal, drawings and tax that never appear on the profit and loss.

Where does the money go if we are making a profit?

The money is normally in one of four places: your receivables ledger, your inventory or work in progress, a payment that has already left for something the profit and loss records slowly or not at all, or a tax liability you collected and have yet to remit. A profit-to-cash bridge from last period's closing balance will show which.

How do I work out which cause is producing our cash gap?

Run an ordered diagnostic rather than reasoning from a list of causes. Reconcile the bank, bridge cash to profit, split working-capital movements into volume and days, measure actual customer payment behaviour, schedule committed outflows, classify forecast variances, and only then investigate margin. The order rules out cheap and common explanations before expensive ones.

What is the difference between a cash flow problem and a profit problem?

A cash flow problem is timing: the money is coming, later than you need it. A profit problem is structural: the economics of what you sell have deteriorated. The test is whether a profit-to-cash bridge and a working-capital decomposition explain the movement. If they do, it is timing. If cash keeps eroding after they are accounted for, it is margin.

Why does my cash flow forecast keep being wrong?

Short-term forecasts miss for five recurring reasons: they forecast revenue instead of collections, they use accounting dates instead of expected payment dates, they omit capital expenditure, loan principal, tax and drawings, they are never compared against actuals, and the systems feeding them carry different data timestamps.

How accurate should a 13-week cash flow forecast be?

There is no reliable published benchmark. The accuracy percentages circulated for 4, 13 and 26-week horizons trace back to other content rather than to studies with disclosed samples and methods. Track your own error instead — variance in closing cash at one, four, eight and thirteen weeks ahead, with inflow and outflow variance measured separately.

Our cash position dropped overnight. How do I investigate it quickly?

Start at the bank portal and identify the transactions that caused the movement, then reconcile your ledger to the bank to rule out a feed or data problem. Check one-off outflows next — tax, loan principal, capital expenditure, drawings or an early supplier run — then which expected receipts did not arrive. A tool that syncs from your accounting platform lags the bank and is not the right first instrument.

Does VAT cause cash flow problems?

UK VAT creates a periodic cliff because the tax point is normally the invoice date, so VAT falls due on invoices customers have not yet paid, and the return and payment are due one calendar month and seven days after the quarter ends. Australian and New Zealand GST and US state sales tax behave differently, so treat each jurisdiction on its own terms.

Can cash flow forecasting software fix a cash gap?

No. It identifies which gap you have and how much time you have to act. Software maps invoices and bills to expected payment dates, consolidates accounts and entities, and makes variance visible weekly. Closing the gap still comes from collections, terms, the timing of committed outflows, financing, or fixing margin.

Who can see our financial data if we use a cash flow forecasting tool?

In Float, access is controlled by role-based permissions with three levels — Admin, Editor and Viewer — so you decide who can view, edit or manage the forecast. Two-factor authentication is mandatory for Xero users. Float reads data from your accounting platform rather than connecting to your bank, so no banking credentials are involved.

Float connects to Xero or QuickBooks Online and gives finance teams a rolling 13-week cash forecast built from your accounting data, synced daily. Start a 14-day free trial with no credit card, or see pricing.