Restaurant cash flow forecast: a 13-week model that reflects how a restaurant actually gets paid

Four timing effects decide whether a restaurant has cash on a Tuesday, and none of them appear on a profit and loss statement: card settlement arrives after the sale, vendors are paid on terms, payroll lands in arrears on its own cycle, and sales tax sits in your account for weeks before leaving in one lump. Model those four and the shortfall stops being a surprise.

Set your assumptions below. The forecast recalculates as you type.

Why the four timing effects matter more than the margin

A restaurant can run a healthy prime cost and still miss rent. The profit and loss statement records a sale on the day it happens and a food cost on the day it is used. Cash obeys a different calendar.

1. Card settlement runs behind the sale

Most of the week's revenue arrives in the bank one to three business days after the guest pays, and a Friday and Saturday settle after the weekend. A week that starts with a large vendor payment and ends with a strong service can still run dry mid-week, because the strong service has not landed yet. Set the settlement option to the following week if your deposits typically clear after the week closes.

2. Vendor terms move the cost, not the purchase

Food bought this week on net 14 leaves the account two weeks from now. That is helpful while volumes are flat and dangerous when they are rising, because a growing restaurant is paying last month's smaller invoices out of this month's larger sales, and the catch-up arrives all at once. The model applies your terms to the week the food was used, so a sales increase shows the payment landing later.

3. Payroll is lumpy in a way weekly sales are not

On a fortnightly cycle, six or seven weeks in a thirteen-week quarter carry no payroll at all and the rest carry double. Operators who think in weekly averages consistently underestimate the weeks that carry a pay run alongside rent.

4. Sales tax is cash you are holding, not cash you have

Every week you collect tax on top of your sales and hold it in the same account as your own money. On a monthly filing cycle that balance builds for four weeks and leaves in a single payment. The forecast shows the held balance separately so you can see how much of your bank balance is already committed. This is the single most common reason an operator believes they are in better shape than they are.

Reading the result

The number that matters is not the closing balance in week thirteen. It is the lowest closing balance across the thirteen weeks, and the week it occurs. A forecast that ends the quarter comfortably but dips below the buffer in week six describes a business that needs a decision in week four.

If the model dips, the levers available in the short term are narrower than they look. Cutting food cost does not release cash for two weeks because you are still paying for food already delivered. Cutting labor releases cash in one week. Moving a vendor from net 7 to net 30 releases roughly three weeks of purchases once, and once only. Delaying a sales tax remittance is not a lever at all, whatever the shortfall.

What a thirteen-week model is for

Thirteen weeks is a quarter, and a quarter is long enough to contain a full seasonal swing, a quarterly tax filing and at least one rent cycle, while staying short enough that the weekly figures are still grounded in what you know rather than what you hope. It is the standard horizon lenders and landlords expect when they ask to see a projection, which is a second reason to keep one current.

Run it weekly, replacing the oldest week with a new week thirteen and putting actual sales into the weeks that have closed. A forecast that is rebuilt from scratch each quarter is a document. A rolling one is a control.

Common questions

What is a 13-week cash flow forecast?

A week-by-week projection of money entering and leaving the bank over one quarter, built from expected sales and the actual timing of collections and payments rather than from accounting periods. It is the standard format for showing a lender, landlord or investor that a business can meet its obligations.

Why is my restaurant profitable but short of cash?

Usually a combination of card settlement arriving after the sale, vendor invoices from a busier prior period landing now, a pay run coinciding with rent, and sales tax leaving in one payment. Each is a timing difference rather than a loss, which is why the profit and loss statement does not show it.

How much cash should a restaurant hold?

There is no single correct figure, but a buffer is normally set against the largest predictable outflow rather than against average weekly costs. Take the worst week in this forecast, which is typically the week carrying rent, a pay run and a tax remittance together, and hold enough to clear it without relying on that week's takings.

Does this tool store or send my figures?

No. The calculation runs in your browser. Nothing is transmitted to us or saved, and closing the page discards everything. The CSV download is generated locally on your device.

Can I use this for more than one location?

Run a separate forecast per location first, because the timing effects differ by site, then consolidate. A group forecast built from a blended average hides the unit that is causing the problem, which is the usual reason a multi-unit operator cannot tell which location is consuming the cash.