Two of the most common training conversations I have with new general managers, and half of the underperforming ones, start with the same misunderstanding. They think their P&L is a single number with a lot of lines under it. It is not. A restaurant P&L is actually three different reports stacked on top of each other, and each report answers a different question with a different management move.
The two most confused lines on that stack are prime cost and controllable cost. They look similar. They live near each other. They move in the same direction most weeks. And they are managed completely differently. Getting the distinction right is worth several points of margin per year at the unit level. Getting it wrong is why capable general managers spend their weekly reviews defending things they were never able to change.
Prime cost: what it is and what it targets
Prime cost is the sum of cost of goods sold and total labor, expressed as a percent of net sales. That is the whole definition. Total labor means everything on the payroll line: wages, payroll taxes, benefits, workers compensation, and any temporary or contract labor billed to the unit.
The formula:
Prime cost % = (COGS + Total Labor) / Net Sales
= (Food purchases +/- inventory movement + Total Labor) / Net Sales
Prime cost is the single most important operating number on a restaurant P&L. If prime cost is healthy, most operations can absorb almost anything else that goes wrong for a quarter or two. If prime cost is broken, no amount of skill on the operating expense lines can save the unit.
Segment benchmarks I have seen hold up across roughly $54M in combined operating scope over 16 years:
- Full-service casual dining: 60 to 65 percent
- Fine dining and chef-driven concepts: 62 to 68 percent
- Fast casual: 56 to 62 percent
- Quick service: 55 to 60 percent
- Coffee and specialty beverage: 45 to 55 percent
- Pizza and delivery-heavy formats: 55 to 62 percent
These are ranges, not points. Where you should sit inside the range depends on labor market wages, real estate cost, day-part mix, and format specifics. But if you are 3 points above the top of your range, you have a fix-it problem, not a benchmark-argument problem.
Controllable cost: what it adds and why it matters
Controllable cost is prime cost plus every operating expense line the unit-level manager can actually move within the reporting period. In most operating structures I have run, that means:
- Supplies (paper goods, chemicals, bar consumables)
- Small wares (plates, glassware, service pieces under a capitalization threshold)
- Uniforms and linens
- Repair and maintenance under a dollar threshold (typically $500 or $1,000)
- Utilities where sub-metered and manager-influenceable
- Credit card fees (in some structures, as a function of pricing decisions)
- Cleaning services if managed at the unit
It excludes rent, insurance, corporate allocations, franchise fees, depreciation, amortization, and any capital charges. Those are structural costs. The general manager does not get to move them and should not be graded on them.
Healthy controllable cost lands between 72 and 78 percent of sales in most full-service formats. The delta between prime cost (~60 percent) and controllable cost (~75 percent) is roughly 12 to 18 percent of sales. That is the operating expense band the general manager can actually work.
Fig. 1 · Prime cost, controllable cost, and everything below the general manager's decision rights.
The management moves are completely different
This is the part most operator training gets wrong. Prime cost and controllable cost do not respond to the same interventions. They move on different time horizons with different tools.
Prime cost moves in days
Labor variance responds to the next schedule you post. If Tuesday's labor was 34 percent and target is 30, Wednesday's schedule reflects the fix. Food cost responds to the next prep list, the next portion check, the next inventory count. The tools are the daily flash report, the schedule, the prep sheet, and the walk-in count. These are weekly at the outside and often daily.
Controllable-only cost moves in weeks or months
The lines that are controllable but not prime move slowly. Supplies. Small equipment. Uniforms. R&M under the threshold. These respond to vendor consolidation, standardization decisions, and capex trade-offs. You cannot cut your paper goods cost in one week. You can renegotiate a paper contract that changes the per-unit price for the next 12 months.
Non-controllable cost moves in quarters or years
Rent. Insurance. Corporate allocations. Franchise fees. Depreciation. These move on real estate cycles, insurance renewals, and corporate policy. The general manager does not have a play here. Making them defend it in a weekly P&L review is a management failure, not a general manager failure.
Never ask the general manager to defend a line they cannot move. It teaches them that the P&L review is theater. Once the review is theater, everything above the line stops moving too.
The example that clarifies it
A real reference case from a $30M Bay Area group. Three underperforming units, weekly P&L reviews had turned into hour-long debates about the rent allocation model and insurance passthrough. The general managers were exhausted. Prime cost was 68 percent versus a 62 percent target. Controllable was 82 percent versus a 76 percent target. Structural cost was fixed and untouchable inside the year.
The change: split the review. Prime cost got a weekly Monday review, 25 minutes, general manager and area director. Controllable-only cost got a monthly review on the first Monday of the month, 45 minutes, general manager and area director together with the operations finance lead. Non-controllable cost got a quarterly conversation with corporate finance, general manager not in the room.
Prime cost dropped from 68 to 61.4 percent inside 90 days. Controllable dropped from 82 to 76.5 percent inside four months. Non-controllable was unchanged, but nobody was arguing about it in the weekly review anymore. Combined annual profit swing across three units, in the ballpark of $1.6M. Almost all of it from the reallocation of general manager attention.
Before: Weekly review agenda After: Weekly review agenda - Sales vs forecast - Sales vs forecast - Food cost (line by line) - Prime cost trailing 7d - Labor cost - Labor % vs demand - Supplies - Food % vs target - Uniforms - Top variance in prime - R&M - One action for this week - Rent (why so high?) - Insurance (unfair) (Controllable review = monthly) - Corporate allocation (Non-controllable = quarterly) - Utilities - One action if there is time Total time: 55-70 min Total time: 25 min Actions taken per week: 0.7 Actions taken per week: 3.2
Where operators most commonly get this wrong
Four failure patterns I see repeatedly:
They put labor in controllable, not in prime
Some P&L templates put labor in the operating expense section rather than in COGS. This is a bookkeeping choice that has no operating meaning. Wherever it sits on the report, labor is part of prime cost. Manage it as prime.
They exclude payroll taxes and benefits from labor
The general manager sees "wages" and calculates prime off wages alone. Real labor cost is 15 to 25 percent above wages depending on jurisdiction and benefit load. Excluding taxes and benefits makes prime look 3 to 5 points better than it is, which sets a target that nobody can hold to the bottom line.
They include structural cost in controllable
Rent gets pulled into "controllable" because it is a fixed dollar amount and the operator likes the way it makes their operating discipline look. Then the general manager gets graded on a percent that moves inversely with sales without any move they can make. Rent is not controllable. Sales strategy is.
They review controllable weekly
The reverse of the mistake above. The weekly P&L review turns into a debate about supplies variance driven by a one-time order timing. Weekly noise on a monthly signal. Move controllable-only lines to a monthly cadence and the weekly review gets its focus back.
What each cadence's review actually looks like
Three different reviews, three different agendas, three different owners. Do not run them as the same meeting.
Weekly prime cost review
25 to 35 minutes. General manager and area director. Every Monday at the same time. Agenda: sales versus forecast, prime cost trailing seven days, labor as a percent of sales with the day-part breakdown, food cost trailing four weeks, top variance in prime, one action for next week. That is the whole meeting. No supplies. No R&M. No rent. The general manager owns the room. The area director asks questions and commits to the one action.
Monthly controllable review
45 minutes. First Monday of the month. General manager, area director, and operations finance lead. Agenda: month-over-month movement on the six controllable-only lines, vendor renewal calendar for the next 60 days, standardization opportunities across units, one committed initiative for the month. This is where you talk about paper goods, uniform providers, chemical vendors, and small equipment plans. It is a slower conversation with real dollar decisions in it.
Quarterly structural review
60 to 90 minutes. Area director, VP of operations, CFO or finance director, and corporate real estate or facilities if applicable. General manager not present. Agenda: rent trends, insurance renewals, corporate allocations, capital plan for the next 12 months. This is a leadership conversation, not a general manager conversation. Excluding the general manager is not a snub. It is a signal that these decisions do not belong on their plate.
How to install the split
You can do this in one week without any technology change. The only work is agenda redesign and role clarification.
- Monday. Rewrite the weekly P&L review agenda to five items. Sales vs forecast, prime cost trailing seven days, labor vs demand, food vs target, top variance in prime. Kill the rest.
- Tuesday. Draft the monthly controllable review agenda. Six items maximum. Supplies, small wares, R&M under threshold, uniforms, cleaning, utilities. Schedule the first Monday of every month.
- Wednesday. Write a one-page structural cost note for the general managers. This is what you own, this is what you do not own, this is where the non-controllable conversation happens.
- Thursday. Meet with the area director. Align on which lines are in which bucket. Kill any debate about it before it hits the general managers.
- Friday. Communicate the change to the general managers. Frame it as clarity, not a downgrade. Most general managers will thank you.
The point
The P&L is not one report. It is a stack of three reports at three cadences with three owners. Prime cost, weekly, general manager. Controllable-only, monthly, general manager with finance. Non-controllable, quarterly, area director and corporate finance. Stack the reports and put each in its own review, and every conversation gets shorter and more effective.
Confusing prime and controllable is why general managers feel exhausted by their P&L reviews and unable to name the number they own. Separating them is the fix. It costs nothing to do, changes almost no numbers on the report, and typically shows up in prime cost within a quarter because the general manager can finally focus on what they can move.
Weekly. Prime cost. General manager. That is the whole game above the line. Everything else has its own room.