The Year-End Close Starts in October

Most operators meet their year-end close in February, when an accountant asks for something they cannot produce. The inventory count nobody wrote down. The vendor statements still in a drawer. A gift card number that turns out to be revenue when it should have been a liability.

By then the year is shut. The answers are estimates, every correction costs money, and the financials that come out the other end are the ones a bank or an investor is about to read.

None of the work below takes long in the fourth quarter. All of it is expensive in February.

An operation that closes on calendar months is comparing a five-weekend August to a four-weekend September and calling the difference performance. It isn’t performance. It’s the calendar.

If you’re moving to a 4-4-5 or a 13-period year, the changeover has to land on a year boundary or it creates a mess that takes two years to unwind. Decide in October. Agree the period-end dates, then load them into the point-of-sale, the payroll calendar and the reporting pack at the same time, because getting two out of three is worse than not starting.

If you’re staying on calendar months, at least build the comparison honestly: sales per operating day, with weekend days per period printed on the face of the report. It costs nothing and it ends a recurring argument.

Food cost calculated from purchases is a purchasing report. Cost of goods sold is opening inventory plus purchases minus closing inventory, and the count on the last day of the year sets both this year’s profit and next year’s starting position.

Three decisions are worth making before the count rather than during it.

Who counts what, and in what order. A count sheet ordered by storage location instead of alphabetically will cut the time roughly in half and take most of the errors out with it.

What gets counted at all. Settle your treatment of opened bottles, prepped items and non-food supplies now and apply the same treatment next year. Consistency matters more here than precision. A method that is slightly wrong but stable still gives you a usable trend. A method that changes every year gives you nothing.

Cut-off. Deliveries arriving on the last day either belong in the count and in payables, or in neither. Both is the most common and most expensive error in the whole exercise, and it is the one an auditor finds first.

Two patterns cause most year-end restatements in this industry.

The first is invoices that never reached accounting. Broadline distributor statements reconcile cleanly because the statements arrive. It’s the small local vendors that go missing. The linen company, the knife sharpener, the equipment repair on a Saturday night. Ask every vendor for a statement in November and reconcile it. It is the fastest way to find cost you didn’t know you had, and it is better to find it in a year you can still plan around.

The second is accruals nobody set up. Payroll for days worked but not yet paid at year end. Accrued vacation, where you carry it. Sales tax collected in December and remitted in January. And gift cards, which deserve their own paragraph: a gift card sold is a liability, not revenue, until it’s redeemed. An operator who books a strong December of gift card sales as income has overstated the year and will be surprised a second time in January when the redemptions land as cost with no matching sale.

The tip layer is where restaurant compliance concentrates, and it is not a January job.

If you take a tip credit, the arithmetic has to hold for every pay period, not on average. Tipped wage plus tips must reach the full minimum wage in each period, and the employer makes up any shortfall. State law stacks on top of that, and some of it is easy to miss: New York, for example, adds spread-of-hours and call-in pay obligations that arise from the shape of a shift rather than the number of hours in it, which is exactly the kind of rule that slips when scheduling lives in one system and payroll in another.

Rates change annually. The mechanics don’t. Check the rates that apply to you before the last quarter’s payrolls run.

If you’re a large food or beverage establishment, Form 8027 reports gross receipts, charged tips and reported tips, with allocation following when reported tips fall short of the threshold. Everything it needs is charged receipts and charged tips by period. That is trivial to pull in October and genuinely painful to reconstruct in February.

And if you’re paying the employer’s share of FICA on reported tips, the FICA tip credit returns some of that. It’s claimed on the return, but it is only ever as good as the tip reporting beneath it. Operators who tighten reporting in the fourth quarter recover more of it, and the ones who don’t are leaving a credit on the table while paying for the payroll that generated it.

Every restaurant has a gap between reported sales and deposited cash. The gap is legitimate: card fees, chargebacks, delivery commissions, marketing deductions, cash tips paid out, and settlement timing across a weekend.

Having the gap isn’t the problem. Not being able to explain it is.

Take one week in October and tie it end to end. Gross sales per the register, less comps and discounts, to net sales. Net sales to the settlement report for every tender type. Settlement to the bank. Whatever breaks in that one week is breaking in all fifty-two.

Third-party delivery deserves its own look. Recording only the net deposit removes both the revenue and the commission from your books, hides what the channel actually costs, and makes it look more profitable than it is. Record it gross, with commissions and fees on their own lines, and you can finally answer whether the channel earns its place.

Equipment, a build-out, a new register system. What was capitalized and what was expensed as repairs changes both the tax position and the balance sheet a lender will read. Walk the asset list before year end, write off what has actually been thrown out, and check that a leasehold improvement isn’t sitting in repairs and maintenance because that’s where it landed the day the invoice arrived.

A budget approved in March governs nine months. A budget approved in December governs twelve.

Build it by period using this year’s actual pattern of trade, not the year divided by twelve. Model prime cost, meaning food, beverage and fully burdened labor, as the control number. It’s the only figure an operator can move inside a single week.

Then hold one line back and treat it as a cost rather than a residual: what the owner takes out. The most common reason a profitable operator can’t fund a second location is that growth consumed every dollar the business made, and the budget never said what the owner was supposed to live on. That isn’t a financing problem. It’s a budgeting one, and October is when it gets fixed.

A closed prior year. An opening balance sheet you’d hand a lender without a covering explanation. A first quarter already being measured against a budget that existed before it started. And an accountant who receives a file rather than a box.

None of that is achieved in January. All of it is decided now.

Sonny Grappone runs Back of House Accounting, an outsourced accounting, payroll and CFO practice working exclusively with restaurants, hotels and hospitality groups. Based in Melville, New York, the firm serves operators in all 50 states.