Labor is capacity, not cost — and every rule of thumb the industry uses to run the labor line treats it as the opposite.
The operator was taught the percentage. Labor should run somewhere between 25 and 35 percent of sales, the exact number depending on who taught it and when, and the operator's job is to hit it. The percentage is not a design. It is a floor set by reference — last year's number, the consultant's benchmark, the number the last operator left behind — and it gets hit the only way a floor can be hit: by cutting the thing the number measures. The industry has been teaching operators to manage labor as a cost line for decades, and the teaching is not wrong about the arithmetic. It is wrong about what the arithmetic is for.
The Percentage Cannot See A Step
The reason the percentage can't be the design is mechanical: capacity does not ramp. It steps. A server can carry about fifty covers before the operation needs another one. A bartender can run about a hundred drinks. The kitchen steps in bigger jumps — a station, a line, a shift. Revenue, meanwhile, ramps continuously — the restaurant sells one more cover, then one more, then one more, and every cover between the steps is sold on capacity the operation does not have yet.
That is the Labor Step-Function: the gap between the staircase of capacity and the ramp of demand, and it is the single most expensive piece of arithmetic in the building. The operator who staffs to the percentage staffs to the middle of a step — either carrying capacity he is not selling, or selling covers he cannot carry. The operator who staffs to the step buys the next server when the ramp crosses fifty, and the fifty-first cover pays for her.
The Inherited Rules
The percentage is not the only inherited rule. The whole labor line runs on four of them, and each one has the same shape: it was true once, in a specific operation, and the industry kept the shortcut and dropped the reason. The percentage target treats the number as the design. Turnover-as-fact teaches the operator that people leaving is weather, to be absorbed, not a signal to be read. The hours dial assumes the only lever is how many hours the people you already have are on the clock. And the meal-credit harvest — the arithmetic that turns a discounted or comped meal into a labor savings — treats the cast's own meals as a line item to be squeezed.
None of these are arguments. They are defaults — positions the industry arrived at when someone stopped asking the question, and they get passed down as counsel because they are easy to teach and easy to audit. The operator who runs a restaurant on them is not making decisions about labor. He is executing a set of instructions written for an operation he does not own, in a market he is not in, by a standard he never set.
Each rule fails the same way: at the mechanism. The percentage cannot see a step. Turnover-as-fact cannot see a design. The hours dial cannot see capacity that is not hours. The meal-credit harvest cannot see the second job the meal dollar does. Every one of them is a lever pointed at the wrong mechanism, and the operator pulls it and wonders why the labor line never actually improves.
The Harvest That Costs More Than It Saves
The meal-credit arithmetic is the clearest example of counsel kept past its reason. Run it for a 25-employee operation and the harvest works out to about $40,495 a year — a number the industry's own cost-controls teaching treats as a savings to be captured. But the meal dollar does two jobs for the same money. The cast member you feed is the cast member who stays, and the cast member who stays is the capacity you do not have to re-buy. Sixty days of tenure is worth roughly a twenty percent reduction in labor cost — the number the same counsel published and then taught operators to ignore in favor of the harvest.
The operator who harvests the credit saves forty thousand dollars and re-hires the cast. The operator who feeds the cast keeps the capacity. Same money, opposite outcomes, and the difference is which question the operation asks.
Turnover Is A Design Signal
The turn on turnover is the whole argument in one move. The default operation absorbs turnover as a cost of doing business and budgets for the re-hire. The designed operation reads the same departures as a signal about the design and fixes the design, because a cast that leaves is a cast that was never held. The difference is not the turnover rate. It is which question the operation asks when someone walks.
The labor line is not a cost to manage. It is capacity to design — and the operator who reads it as a cost will spend his career buying back, at re-hire prices, capacity he never had to lose. Labor is capacity. The schedule is where you decide how much of it the restaurant will have.
Read the prosecution: https://hacksterism.jeffreysummers.com/the-labor-line-runs-on-rules-of-thumb/
Read the architecture: https://physics.jeffreysummers.com/labor-is-capacity-not-cost/
Read the rework: https://jeffreysummers.com/the-labor-read/
One move this week: stop reading the percentage and read the steps. Take the last four weeks of covers by daypart and find where the ramp crossed the steps — the night the fiftieth cover walked in, the shift the hundredth drink was poured. That is where the schedule was wrong, and the percentage never told you, because the percentage cannot see a step.
— Jeffrey
Digging Deeper: The Labor Step-Function, Cast Design, The Dead Zone Is a Choice — definitions in the Knowledge Base at https://kb.jeffreysummers.com/. More from Restaurant Physics at https://physics.jeffreysummers.com/, more from Hacksterism at https://hacksterism.jeffreysummers.com/, and the hub read at https://jeffreysummers.com/the-labor-read/.
SOURCES CITED
Pavesic, David V., Labor Cost, Restaurant Manager's Pocket Handbook Series, Lebhar-Friedman Books, 1999.