How Many Employees Should I Schedule Each Shift at My Steakhouse?

The two ways steakhouses actually decide headcount

Almost every steakhouse staffs its shifts using one of two methods, and they produce very different schedules from the same room.

The first is the habit method, sometimes dressed up as "experience." The general manager looks at last week's grid, copies it forward, adjusts for who requested off, and publishes. The counts came from somewhere years ago — the previous GM, the opening crew, a busy December that never repeated — and they persist because nobody has a reason to change them. Habit scheduling is fast, costs nothing, and is invisible until you audit it. Its failure mode is symmetrical and expensive: you carry two extra servers through a dead Tuesday and you run a Saturday with one host short and a 40-minute quote on a 25-minute wait. Both errors are silent. The overstaffed Tuesday shows up only as a labor percentage nobody drills into; the understaffed Saturday shows up as a two-star review about slow service that gets blamed on the kitchen.

The second is the gross-profit division method. You agree on a single number — the gross profit one average employee should produce working an average shift — then divide each shift's actual trailing gross profit by that number to get headcount. The output is a count per shift per day of week, derived from receipts rather than memory. It costs you an afternoon of data pulling and a leadership conversation about what the target should be, and it produces a schedule you can defend to a server who thinks they got cut unfairly, because the number came from the shift, not from a manager's preference.

The practical difference on a 120-seat steakhouse doing $3.2M a year is usually two to four bodies per week that habit was carrying without justification, plus one or two shifts a week that habit was chronically underserving. At $18 to $22 an hour loaded for a six-hour shift, three redundant weekly shifts is roughly $350 a week, or $18,000 a year, sitting inside a schedule nobody thought was broken. The understaffed side is harder to price but generally larger, because a steakhouse loses a $180 four-top to a bad wait far faster than it loses $115 in labor.

There is a third option people reach for — the flat labor-percentage rule, "keep labor under 30%." That is a scoreboard, not a scheduling method. It tells you at the end of the week whether you were right; it does not tell you how many people to put on Thursday dinner. Use it as a check on the division method's output, never as a substitute for it. RevOps discipline in a restaurant means the same thing it means anywhere else: the plan comes from a model, and the percentage is how you audit the model.

How to decide which method your room needs

The honest answer is that the division method wins in almost every steakhouse, but the transition path depends on what data you already have and how much variance your room carries week to week. Work the decision in this order.

Do you have shift-level sales data? If your POS is Toast, Square, Aloha, Micros, or anything modern, you have hourly sales exports going back as far as you have been open. If you can pull a report that shows net sales by day part, you can run the division. If you genuinely cannot — an old system, a recent ownership change, missing months — start tracking now and use an estimate in the meantime, described in the numbers section below.

How much does week-to-week variance move you? Compute the coefficient of variation on the same shift across twelve weeks: standard deviation divided by mean. Under about 15% and a single fixed count per shift will hold most weeks. Above 25% — common for steakhouses near a stadium, a convention center, or a seasonal resort — a fixed count will be wrong in both directions, and you need the division method plus a call-in or on-call tier layered on top.

Is anyone contesting the schedule? If servers are arguing about who gets Friday, or if you have had a turnover spike among the people who consistently get the slow shifts, habit scheduling is already costing you retention. Division gives you a neutral answer.

The decision loop matters more than any single run of it. The first division you do will be somewhat wrong, because your first target number is a guess informed by your margins rather than a measured fact. Four weeks of running it, checking labor percentage and service quality, and adjusting the target is what converts the guess into a number you can trust for a year.

The numbers behind each option

Setting the per-employee target. The target is the gross profit one average employee should produce on an average shift. For a steakhouse, $320 per shift is a reasonable opening figure, and here is why it lands there rather than at a fast-casual number. A steakhouse server carries fewer tables — typically three to five in a full-service room versus eight or more in a casual concept — but each table is worth far more. At a $70 average check per guest and a 3.5-person average party, a four-table station turning twice across a dinner shift moves roughly $1,900 in sales. Steakhouse food cost usually runs 32% to 38% because protein is the center of the plate, so gross profit on that food is roughly 62% to 68%. Beverage, and particularly wine, runs a much better margin — often 65% to 75% gross profit — and in a steakhouse beverage frequently accounts for 30% to 40% of the check. Blend those and you are looking at a gross profit in the neighborhood of 65% of sales before labor.

Apply that: $1,900 in server-station sales at 65% gross profit is about $1,235 in gross profit generated at that station. But the station does not run itself — a busser, a share of the host, a share of the line, and a share of the dish pit all sit behind it. Distributing that gross profit across the full complement of hourly bodies is what pushes a per-employee figure down into the low hundreds. Landing on $320 as a floor means an average shift with an average guest count and average service clears the number without heroics, and a good shift beats it.

Running the division. Pull trailing three to six months of gross profit, bucketed by day of week and day part. A realistic mid-size steakhouse pattern:

• Monday dinner: $1,280 gross profit → 4 people
• Tuesday dinner: $1,280 → 4
• Wednesday dinner: $1,920 → 6
• Thursday dinner: $2,560 → 8
• Friday dinner: $3,840 → 12
• Saturday dinner: $4,160 → 13
• Sunday dinner: $2,240 → 7
• Weekday lunch (if you serve it): $960 → 3

That is 57 shift-bodies a week for dinner plus lunch coverage, and it is a genuinely different shape than the flat "eight on, every night" that habit produces — which would be 56 bodies distributed exactly wrong, four too many on Monday and four too few on Friday.

Splitting the count across roles. The division gives you a total; the room gives you the mix. On a twelve-person Friday dinner in a 120-seat steakhouse, a workable split is five servers, two bussers, one host, one bartender, and three back of house on the line, with the dish position and the kitchen manager budgeted separately as fixed overhead rather than counted against the service target. Most operators exclude salaried managers and dedicated dish from the per-employee math entirely, because those roles do not scale with covers the way service roles do, and folding them in muddies the target. If you do include them, your target drops — roughly to $250 to $270 on the same volumes — and you must apply the lower number consistently or the division stops meaning anything.

Placing bodies against the receipt curve. The count answers how many; the hourly sales export answers when. A typical steakhouse dinner posts 15% of its tickets between 5:00 and 6:00, 30% between 6:00 and 7:30, 35% between 7:30 and 9:00, and the remainder trailing to close. That means your twelve-person Friday is not twelve people from 4:00 to 11:00. It is four on at 3:30 for prep and setup, six more staggered in between 5:00 and 6:00, the last two landing at 6:30 for the peak, and cuts starting around 9:15 as the dining room drains. Staggered starts against the receipt curve routinely save 8% to 12% of scheduled hours versus block scheduling the same headcount, without touching the count itself.

The cost of being wrong. At a $19 loaded hourly rate on a six-hour shift, each unnecessary body costs $114. One extra person on every Monday, Tuesday, and Sunday dinner is $342 a week, $17,784 a year. On the other side, an understaffed Saturday that pushes your wait quote past 45 minutes and walks four four-tops costs roughly $980 in sales, of which about $637 was gross profit, in one night — more than five weeks of the overstaffing error. This asymmetry is the single most important thing to understand about the math: be precise on slow shifts, and round up on peak shifts.

Implementation and sequencing

Rolling this out badly gets you a mutiny among servers who lose shifts, so sequence it deliberately over about six weeks.

Week one — agree on the target in a room with your leadership. Do not set it alone and announce it. Pull your P&L, compute your actual blended gross profit percentage from food and beverage, and work backward with your GM and chef to a number everyone believes. Then say it plainly to the whole crew: in this steakhouse, an employee who shows up, covers an average station, and gives average service should produce no less than $320 a shift in gross profit. Framed as a floor rather than a quota, it becomes a yardstick people can use to argue for more hours instead of a stick used to cut them.

Week two — pull and clean the data. Export gross profit by day part and day of week for the trailing three to six months. Strip the anomalies before you average: the New Year's Eve prix fixe, the two nights the walk-in died, the week the road was closed. Those events belong in a separate holiday plan, not in your baseline. Six months is better than three because it smooths a slow February; if a menu price increase landed mid-window, weight the post-increase weeks more heavily or restart the window there.

Week three — build the grid and compare it to what you actually ran. Divide, get your counts, then put them side by side with last quarter's published schedules. The gaps are the finding. Expect to see two to four bodies of weekly overstaffing on early-week dinners and one chronically thin peak shift. Do not publish yet.

Week four — publish with the peak shifts corrected first. Add to the understaffed Friday and Saturday before you subtract from anywhere. Fixing the expensive error first buys you goodwill and shows the crew the method gives hours as well as takes them.

Weeks five and six — trim the slow shifts through attrition and cross-training, not cuts. Reduce Monday and Tuesday counts by not replacing the next departure, and by cross-training a server to bartend so one body covers two roles on a thin night. Then check labor percentage and service metrics — wait quote accuracy, table turn time, complaint volume — before locking the pattern.

Tooling. Any restaurant scheduling product will publish the grid — 7shifts, Homebase, When I Work, Deputy, Sling, Connecteam, HotSchedules, Workforce.com, and Findmyshift all do it, and several connect to a POS so you can watch labor against sales during the shift. None of them will tell you the Friday rush needs twelve people. You bring the count from the division; the software handles publishing, swaps, clock-in, and compliance. Pick on pricing shape: per-location pricing suits a steakhouse with a large part-time roster, per-user pricing suits a lean stable crew. Check current pricing on each vendor's own site before you commit, since plan tiers change.

Compliance and edge cases. If you operate in a jurisdiction with predictive-scheduling or fair-workweek rules, publishing changes on short notice carries penalty pay, so lock the new pattern before the notice window rather than adjusting live. Watch overtime as you consolidate — cutting a Monday shift and giving those hours to someone already at 36 can cost more than it saves. And keep an on-call tier for high-variance rooms: two people who know they may be called by 3:00 p.m. costs nothing and covers the expensive side of the error.

Recalibration. Revisit the target quarterly, and immediately after any menu price change, minimum wage change, or significant shift in food cost. A 10% menu price increase raises gross profit per cover, which means the same shift now supports the same headcount at a higher target — or the same target at a lower headcount. Decide which lever you want deliberately rather than letting the schedule drift back toward habit.

Related questions

What is a good labor cost percentage for a steakhouse?

Full-service steakhouses commonly target total labor in the high 20s to low 30s as a percentage of sales, with back of house heavier than casual concepts because of butchery and line work. Use it to audit the division method's output, not to set headcount.

Should I schedule the same count for lunch and dinner?

No. Lunch typically generates a fraction of dinner's gross profit — often a quarter to a third — so a weekday lunch might need three people against a Friday dinner's twelve. Run the division separately for every day part.

How do I staff a holiday or a private event?

Pull the same date from prior years rather than using your baseline. With no history, take your busiest normal weekend count and add 20% to 30%, then cut early if the room is slower than expected. Cutting is cheaper than scrambling.

Do managers and dishwashers count toward the per-employee target?

Most operators exclude salaried managers and dedicated dish, budgeting them as fixed overhead, because those roles do not scale with covers. If you include them, lower the target consistently — otherwise the division stops being comparable across shifts.

How long does it take to see the savings?

Peak-shift corrections show up in service metrics within two weeks. Slow-shift savings take six to twelve weeks because you are trimming through attrition and cross-training rather than cutting hours from existing staff.

FAQ

What if my steakhouse's gross profit per employee is higher or lower than $320?

The $320 figure is a starting point for a mid-range full-service steakhouse, not a universal constant. Your real number depends on your average check, your food and beverage mix, and which roles you include. Compute your own by taking total gross profit for a shift and dividing by the number of hourly service employees who worked it, then averaging across several months. A high-end room with a $110 average check and a deep wine program will land well above $320; a lower-priced room will land below it.

What if I don't have three to six months of gross profit data by shift?

Start logging it immediately, then bridge with an estimate. Take expected covers for the shift, multiply by average check, multiply by your blended gross profit percentage, and divide by your target. If you expect 60 covers at a $70 average check with 65% gross profit, that is $2,730 in gross profit, or roughly eight people at a $320 target. Replace the estimate with real data as the weeks accumulate.

How do I handle a shift where guest counts swing wildly week to week?

Schedule the floor count from the division and add an on-call tier. Compute the coefficient of variation across twelve weeks of the same shift; above 25%, a single fixed number will be wrong in both directions. Two on-call staff who confirm by mid-afternoon cost nothing on quiet weeks and protect the expensive side of the error on busy ones.

Won't cutting the slow shifts hurt morale?

It will if you cut people. Do it through attrition and cross-training instead — do not replace the next departure, and train a server to cover the bar on thin nights so one body handles two roles. Also correct the understaffed peak shifts first, so the crew sees the method adding hours before it removes any.

How often should I recalculate the target?

Quarterly at minimum, and immediately after a menu price change, a wage change, or a meaningful move in food cost. A 10% menu price increase raises gross profit per cover, which either lets the same shift run leaner or lets the same crew clear a higher target. Pick which one you want on purpose.

Does scheduling software calculate this for me?

No product hands you the headcount. Several connect to your POS and forecast sales or flag labor percentage in real time, which helps you check the plan, but the per-employee gross-profit target is a decision you and your leadership make. The software's job starts after the count exists: publishing, swaps, clock-in, overtime alerts, and predictive-scheduling compliance.

Sources

• https://restaurant.org/research-and-media/research/ — National Restaurant Association research and industry operations reports
• https://www.bls.gov/oes/current/naics4_722500.htm — BLS Occupational Employment and Wage Statistics for restaurants
• https://www.dol.gov/agencies/whd/flsa — U.S. Department of Labor, Fair Labor Standards Act overview
• https://www.dol.gov/agencies/whd/flsa/tips — DOL guidance on tipped employees and tip pooling
• https://www.sba.gov/business-guide/manage-your-business/hire-manage-employees — SBA guidance on hiring and managing employees
• https://www.7shifts.com/blog — 7shifts restaurant labor and scheduling resources
• https://www.deputy.com/ — Deputy demand-based scheduling and compliance features
• https://joinhomebase.com/ — Homebase scheduling, time tracking, and pricing
• https://www.wheniwork.com/ — When I Work shift scheduling documentation
• https://pos.toasttab.com/blog — Toast restaurant operations and labor cost resources

Related on PULSE

• [How Many Employees Should I Schedule Each Shift at My Thrift Store?](/knowledge/q16019)
• [How Many Employees Should I Schedule Each Shift at My Vintage Clothing Store?](/knowledge/q16018)
• [How Many Employees Should I Schedule Each Shift at My Record Store?](/knowledge/q16017)
• [Should I open or buy a Black Angus Steakhouse franchise in 2027?](/knowledge/q14689)
• [Should I open or buy an Outback Steakhouse franchise in 2027?](/knowledge/q14688)
• [Should I open or buy a LongHorn Steakhouse franchise in 2027?](/knowledge/q14686)