Restaurant Cash Flow: Why Profitable Restaurants Run Out of Money

Profit is a judgement about a period. Cash is a fact about a date. Here is where the two come apart, why growth makes it worse, and how to see the tight week before it arrives.

The gap nobody sees until it is too late

The most disorienting conversation in hospitality goes like this: the accounts show a good quarter, the operator knows trade has been strong, and yet there is not enough in the account to pay the rent. Nothing has been stolen and nothing has been miscounted. The business is profitable and it cannot pay its bills, and both statements are true at the same time.

This happens because a profit and loss account and a bank statement measure different things. The P&L matches revenue to the costs of earning it, in the period it was earned. The bank statement records money moving on the day it moves. Everything that sits between those two — supplier terms, payroll dates, tax accruals, loan principal, stock on the shelf, deposits paid on equipment — is a place where profit and cash can separate.

Consider a site doing $32,000 a week with a 60% prime cost and $6,500 of weekly fixed costs. It makes about $6,300 a week, or roughly $82,000 across a quarter. Genuinely healthy. Now add the things the P&L does not show as costs in that quarter: $18,000 of loan principal, a $12,000 tax payment covering the previous quarter's trading, $9,000 of owner drawings and $6,000 of extra stock bought to support a new menu. That is $45,000 of cash gone against $82,000 of profit, and it is entirely normal. The business is fine. The bank balance simply does not look like the profit figure and never will.

Where the money actually goes

There are four routine leaks between the profit line and the bank balance, and none of them is a sign of anything wrong.

Loan principal

Only the interest on a loan is a cost on your P&L. The principal repayment is a reduction of debt, so it appears nowhere in your profit figure while leaving your account every month like everything else. An operator with $4,000 of monthly repayments, of which $600 is interest, has $3,400 a month of cash outflow that their profit statement is silent about. Over a year that is $40,800 of profit that never becomes money you can spend.

Tax on a delay

Sales tax and VAT are collected week by week and paid quarterly, which means for most of the quarter you are holding money that is not yours in an account that looks like it is. Income or corporation tax is worse, because it is often paid on a period that closed months ago, out of cash generated by trading that has nothing to do with it. Both are entirely predictable and both catch people out.

Stock

Every dollar of inventory is a dollar of cash converted into something you cannot spend. Building a walk-in full of stock to support a menu launch is an investment that shows up nowhere on the P&L until the stock is sold, but it hits the bank immediately. This is why inventory turnover is a cash metric as much as a control one, and why cutting stock levels is the fastest source of cash most restaurants have available. There is more on the mechanics in our guide to inventory turnover.

Timing

Supplier terms decide when the food you have already cooked is paid for. On fourteen-day terms, this week's stock leaves the account in a fortnight, which means you are permanently carrying two weeks of purchases as free working capital. Lose those terms and you lose that capital overnight, without a single number on the P&L changing.

Growth is the most expensive thing you can do

The counter-intuitive part is that growing restaurants run into cash trouble more often than flat ones. Growth consumes cash before it produces any, and the faster it happens the wider the gap gets.

A site adding 20% to its sales has to buy 20% more stock, and it buys that stock before the extra sales are banked. It rosters more hours, and those hours are paid in the week they were worked. If it is expanding into a second service or a new channel, there is equipment, marketing and training to pay for — all of it upfront, all of it in cash, against revenue that arrives over the following months.

The same trap sits inside delivery and catering expansion. Delivery revenue arrives after the platform's settlement cycle and after its commission, while the food and labour behind it are paid on the normal schedule — worth checking against your real commission cost before scaling it. Catering is the opposite and much kinder to cash, because you can take a deposit that funds the ingredients before you buy them.

None of this argues against growth. It argues for funding it deliberately rather than discovering halfway through that the growth was being financed by not paying suppliers.

The twelve-week rolling forecast

The fix for all of this is unglamorous and takes about twenty minutes a week. Build a twelve-week forecast in weekly buckets, and update it every week.

Twelve weeks is the right horizon because it is long enough to catch a quarterly payment while you still have options, and short enough that your sales assumption remains defensible. Weekly buckets rather than monthly ones matter because monthly averaging hides exactly the problem you are looking for: a payment on the 3rd and a settlement on the 8th nets to nothing across a month and is a bounced payment in reality.

What goes in each week: expected sales, supplier payments falling due based on your actual terms, payroll on its real dates, fixed outgoings, and every lumpy payment diarised for the year ahead. What comes out is a closing balance per week, and the number that matters is not the balance at week twelve but the lowest balance at any point along the way.

Then update it. Each week, replace the projection for the week that has passed with what actually happened, add a new week at the far end, and note where you were wrong. Within a month you will know whether you habitually over-forecast sales, and by roughly how much, which makes every subsequent forecast better.

Levers that move cash without moving profit

Because the problem is timing rather than trading, most of the fixes cost nothing in margin:

When it is not a timing problem

There is one case where none of the above helps, and it is important to identify it quickly. If your forecast goes negative with no one-off payment in it — if ordinary weeks consume more cash than they generate — you do not have a cash flow problem. You have a trading problem wearing a cash flow costume.

Extending terms in that situation buys weeks and makes the eventual position worse, because you arrive at the same place owing more. The test is simple: check where you sit against break-even, and check whether your prime cost is where it needs to be for your rent and volume. Our break-even analysis guide covers how to work that out properly.

If the trading is sound and the timing is the issue, forecasting solves it. If the trading is not sound, forecasting tells you that honestly and early, which is the more valuable answer even though it is the less welcome one.

Frequently asked questions

What is the difference between profit and cash flow?

Profit measures whether your trading created value over a period, matching revenue to the costs of earning it. Cash flow measures money actually entering and leaving your account on specific dates. Loan principal, tax timing, stock purchases and supplier terms all move cash without matching movements in profit, which is why the two figures rarely agree.

How much cash should a restaurant keep in reserve?

Four to six weeks of total outgoings is a reasonable target for an independent site. That covers a poor month, an equipment failure and a late payment arriving in the same period. Below two weeks, ordinary operational problems become financing decisions, which is where costs escalate.

Why does my accountant say I made money when my account is empty?

Both are usually correct. Check four things in order: loan principal repaid, tax paid on earlier periods, increases in stock holding, and owner drawings. Together these routinely account for half of a quarter's profit and none of them reduce the profit figure your accountant is reporting.

Should I forecast weekly or monthly?

Weekly. Monthly forecasting averages away the exact problem you are trying to see. A month can net to positive while containing a week where a rent payment, a payroll run and a supplier settlement land within four days of each other, and the month view will never show you that.

Do longer supplier terms actually help?

Yes, materially and permanently. Moving from paying on delivery to fourteen-day terms leaves two weeks of purchases in your account indefinitely — on a site buying $9,000 a week, around $18,000 of working capital at no cost. The condition is that you then pay reliably on the agreed date, because terms are withdrawn faster than they are granted.

Can growth cause a cash crisis in a profitable business?

It is one of the most common causes. Growth requires more stock, more labour hours and often equipment or marketing, all paid before the additional revenue is banked. The faster the growth, the larger the funding gap, which is why expansion should be planned with a cash forecast rather than a profit projection.

What is the first thing to do if next month looks tight?

Move dates before you move money. Ring the landlord, the tax authority or the equipment supplier while the problem is still weeks away and ask about payment timing. Early conversations produce arrangements; late ones produce penalties and lost credit terms, which make the following month harder as well.

Run the numbers

Use the free Restaurant Cash Flow Forecast Calculator to apply everything above to your own figures.

Restaurant Cash Flow Forecast Calculator All guides All calculators