Restaurant Cash Flow Forecast Calculator
Enter your sales, payment terms and fixed outgoings to project your bank balance week by week for the next twelve weeks — and find the week it gets tight before it arrives.
Why profit and cash are not the same number
A profit and loss account tells you whether the trading you did was worth doing. It tells you nothing about whether you can pay Thursday's invoices. The two answers come apart because the P&L records revenue and costs in the period they were earned and incurred, while your bank account moves on the day money actually lands or leaves. Sitting between those two moments are supplier terms, card settlement delays, payroll dates, quarterly rent, loan principal and a tax bill that has been accruing silently for three months.
Take a site turning over $32,000 a week at a 60% prime cost with $6,500 of weekly fixed outgoings. On paper it earns roughly $6,300 a week — a genuinely healthy operation. Now put a $12,000 quarterly tax payment in week six and start it with $8,000 in the bank. The quarter is still profitable and the account still goes negative in week six. Profit is a judgement about a period. Cash is a fact about a date, and the date is what your suppliers care about.
This calculator runs that arithmetic forward twelve weeks. It takes your opening balance, projects sales on your current trend, pays suppliers on the terms you actually have rather than in the week you cooked the food, pays labour in the week it was worked, deducts fixed outgoings every week, and drops a one-off payment into whichever week it falls. What comes back is the balance at week twelve, the lowest point along the way, and the gap between the profit you made and the cash you kept.
The timing gaps that create the problem
Supplier payment terms
Terms are the largest single lever on restaurant cash, and the most commonly ignored. On fourteen-day terms, the food you cook in week six is paid for in week eight. That lag is worth two weeks of purchases sitting in your account as working capital — on a site buying $9,600 of stock a week, roughly $19,200 of breathing room you would not otherwise have. Move from cash on delivery to fourteen days and you have effectively given yourself an interest-free facility without asking a bank for anything.
The same mechanic works against you in reverse. Trade dips and your sales fall immediately, but for the next two weeks you are still settling invoices generated at the old volume. This is why a bad fortnight hurts the bank account after it has stopped hurting the P&L.
Payroll runs on its own calendar
Wages do not wait for terms. They leave on a fixed date regardless of what trade did that week, which makes labour the least flexible line in the forecast. Fortnightly payroll produces two months a year with three pay runs, and those months look like a crisis to anyone forecasting on a monthly average. Weekly buckets show them for what they are: a known, dateable event you can plan around.
Card settlement and deposits
Card takings typically clear one to three working days after the sale, which for a Friday and Saturday means the weekend's money arrives on Tuesday. Over a normal week this washes out, but it matters at month end and it matters badly if a settlement run falls after a payment date. Deposits work the other way and are the rare case where cash arrives before revenue — taking a deposit on a large booking or a catering job funds the ingredients before you buy them.
Lumpy payments
Quarterly rent, sales tax or VAT, annual insurance, licence renewals, equipment repairs and loan principal all share one property: they never appear on a weekly P&L in a form that warns you. Loan principal is not a cost at all in accounting terms, yet it leaves the bank every month like everything else. These are the payments that turn a comfortable forecast into an emergency, which is why the calculator gives them a week of their own.
How the twelve-week projection is built
Twelve weeks is the useful horizon. Shorter than that and you cannot see a quarterly payment coming with enough time to do anything; longer and the sales assumption becomes fiction. Each week is treated as a bucket with money coming in and money going out, and the closing balance of one week becomes the opening balance of the next.
Purchases are assumed to track sales at your cost of goods percentage, then shifted forward by your payment terms. A fourteen-day term becomes a two-week lag, so the payment leaving in week eight is for the stock bought in week six.
The sales trend compounds weekly, so a 0.5% weekly figure grows sales by about 6% across the quarter. Set it to zero if trade is flat, and use a negative number if it is declining — a forecast that only models growth is worse than no forecast at all, because it produces confidence without evidence. If you are unsure what to enter, look at your like-for-like sales trend rather than headline revenue.
Reading the low point
The closing balance at week twelve is the least interesting output. What matters is the lowest weekly balance in the period, because that is the moment something breaks. Compare it against your average weekly outgoings to convert it into something meaningful.
| Low point, in weeks of outgoings | What it means and what to do |
|---|---|
| Below zero | A payment will fail on a known date. Act now, while you still have options that do not cost money. |
| Under 2 weeks | One bad week or one late payment away from trouble. No room for an equipment failure. |
| 2–6 weeks | Workable for an established site with predictable trade. Build the buffer before adding commitments. |
| Over 6 weeks | Comfortable. This is the position from which you can negotiate on price rather than on urgency. |
Treat these as orientation, not targets. A site with heavy seasonality needs a far deeper buffer than one with steady weekday trade, because its low point in the model is not its low point in the year. If your trade halves in a quiet season, run the forecast again with that season's sales figure before deciding you are comfortable.
What to do when the line goes negative
The value of forecasting is the time it buys. A shortfall spotted six weeks out has cheap solutions; the same shortfall spotted on the day has expensive ones. In rough order of what to try first:
- Move the payment, not the money. Most one-off bills have some date flexibility if you ask before they are due rather than after. Ringing a landlord in week two about a week six problem is a conversation. Ringing in week six is an incident.
- Ask suppliers for terms. Going from cash on delivery to fourteen days releases two weeks of purchases permanently. Suppliers grant terms to operators who ask early and then pay on the day they promised.
- Reduce stock, not orders. Every dollar sitting on a shelf is a dollar not in the bank. Tightening par levels and improving inventory turnover converts stock into cash once, and it is usually the fastest few thousand dollars available.
- Take deposits on anything bookable. Large parties, events and catering work should fund their own ingredients. This is normal practice and almost nobody objects to it.
- Delay discretionary spend, precisely. Not a vague freeze — a list of specific items moved to specific later weeks, checked back into the forecast so you can see whether it was enough.
- Fix the trading problem if there is one. If the forecast is negative with no one-off payment in it, this is not a timing issue. Check your prime cost and your break-even point, because no payment schedule saves a site trading below break-even.
- Talk to the bank before you need to. Facilities are granted on the strength of a forecast presented calmly in advance and refused when presented urgently after a bounced payment.
Running it as a rolling forecast
A forecast built once is a document. A forecast updated weekly is a management tool, and the difference in value is enormous. The discipline is small: once a week, drop the week that has passed, add a new week twelve, and replace the projected figures for the completed week with what actually happened.
- Update on the same day each week. Monday morning works because you have the weekend's trade and the week's invoices in front of you.
- Record the variance. Where the actual week differed from the projection, note why. After a month you will know whether your sales assumption is optimistic and by how much, which is worth more than any single forecast.
- Keep one version. Multiple scenarios in multiple files is how forecasting dies. Run one base case and stress it by changing one input at a time.
- Stress the sales line, not the costs. Costs are reasonably predictable. Ask what a 15% drop in sales does to your low point, because that is the shock that actually happens.
- Diarise the lumps for the whole year. Tax dates, insurance renewals, rent quarters and licence fees are all knowable in advance. There is no excuse for one arriving as a surprise.
The operators who never have a cash crisis are rarely the most profitable ones. They are the ones who knew about week six in week one, and had five weeks to make a boring decision instead of an expensive one.
Frequently asked questions
Can a profitable restaurant really run out of cash?
Routinely, and it is one of the most common ways a viable business fails. Profit is measured over a period; payments happen on dates. A quarter that earns $80,000 of profit can still leave you unable to pay a $12,000 tax bill in week six if the profit has not converted to cash yet, or has already gone out as loan principal, stock or owner drawings.
Why twelve weeks rather than twelve months?
Twelve weeks is long enough to see a quarterly payment coming with time to act, and short enough that the sales assumption is still credible. Annual cash forecasts for restaurants are mostly guesswork past the first quarter, because a single menu change, a competitor opening or a bad season moves the numbers more than the forecast's precision.
What should I enter for supplier payment terms?
The average across your suppliers, weighted roughly by spend. If your main food supplier is on fourteen days and your drinks account is on thirty, but food is three quarters of your spend, enter something close to eighteen. Enter 0 if you pay on delivery, which is the worst position for cash and the first thing to renegotiate.
Should loan repayments go in fixed outgoings?
Yes, the full repayment including principal. This is exactly where profit and cash diverge: only the interest portion appears on your P&L, but the whole payment leaves the bank. Forecasting on the interest alone is one of the most reliable ways to be surprised by your own balance.
How do I handle a seasonal business?
Do not use an annual average weekly sales figure. Run the forecast twice, once with your quiet-season weekly sales and once with your peak, and plan around the quiet result. The low point that matters is the one in your worst twelve weeks, not the one in a typical twelve.
Does this account for card settlement delays?
Not explicitly, because over a full week the delay largely cancels out. If your settlement runs longer than three days, or a large payment date falls immediately after a weekend, reduce your opening cash figure by two or three days of takings to build the delay into the projection.
What is a sensible cash buffer to hold?
Enough to cover the low point plus a genuine shock. For most independent sites that means four to six weeks of total outgoings held in reserve, which covers a bad month, an equipment failure and a late payment arriving together. Below two weeks, every small problem becomes a financing decision.