How Many Employees Should I Schedule Each Shift at My Cooking School in 2026?

Why the seats-per-instructor ratio decides everything else

Every staffing decision at a teaching kitchen traces back to one number: how many students a single chef-instructor can supervise at live stations without quality or safety slipping. Get that number right and the schedule writes itself. Get it wrong and you either burn payroll on idle bodies or run a class where six students are waiting for someone to check their brunoise while a seventh has a smoking pan.

The ratio is not a preference. It is a function of three physical constraints in the room. First, station count — you cannot teach more students than you have burners, boards, and workspace for, and a 12-station kitchen caps you at 12 hands-on seats no matter how good your instructor is. Second, supervision distance — an instructor working a U-shaped or double-galley layout can visually sweep ten stations in a few seconds; spread those same ten across two rooms and the sweep becomes impossible. Third, intervention frequency — the more often a technique requires the instructor to physically step in (hand-over-hand knife correction, tempering chocolate, checking doneness), the fewer students one person can carry.

Those constraints produce a ratio band rather than a single number. Ten students per chef-instructor is a defensible default for standard hands-on classes: pasta, weeknight dinners, regional cuisine, basic baking. Demonstration-format classes where students watch and taste rather than cook can run considerably higher — 16 per instructor is workable, and larger for pure lecture-and-taste formats, because nobody is holding a knife or standing over an open flame. Advanced technique classes pull the number down. Knife skills, laminated dough, sugar work, and anything involving a deep fryer or a mandoline reasonably cap at 8 per instructor, because the intervention frequency is high and the consequence of an unwatched mistake is a laceration or a burn.

Kids' and teen classes deserve their own line. Supervision load per participant is materially higher, minors are less predictable around hot surfaces, and many jurisdictions impose additional requirements on programs serving minors. Most operators who run youth programming land somewhere between 6 and 8 participants per adult, and many add a second adult in the room regardless of headcount for basic supervision and liability reasons. Whatever number you settle on, write it down, because "we usually put two people on kids' class" is not a number and cannot be scheduled against.

The reason to set the ratio explicitly, with your head chef, in a conversation you can point back to, is that it removes argument from the schedule. When the ratio is written on the wall, a light Wednesday genuinely staffs one instructor and nobody reads it as a slight. When the ratio is a vibe, every schedule becomes a negotiation, and negotiations get resolved by whoever complains loudest — which is how schools end up with three people on an eight-student class and one exhausted chef running twenty.

One more thing the ratio buys you: a defensible answer to enrollment pressure. When a regular asks you to squeeze one more seat into a sold-out class, the ratio tells you exactly what that costs — either it fits under the cap or it triggers a second instructor, and a single extra seat almost never pays for a second instructor's shift. The ratio turns a favor into arithmetic.

The step-by-step process for building the shift

The method is four steps, run once per class block on the calendar, then rolled forward weekly with adjustments.

Step one: fix the per-instructor cap by class type. Sit with your head chef and assign a cap to each class format you actually sell — not a generic number for the whole school. A realistic set might read: hands-on standard 10, hands-on advanced 8, demonstration 16, kids 6, private/corporate buyout per the contract. Write it on a laminated card in the office. Revisit it quarterly or whenever you add a format.

Step two: pull expected seats per block and divide. Do not use capacity, and do not use last week. Use a trailing four-to-six-week average of *actual* attendance for that specific class slot, because Saturday 11 a.m. behaves nothing like Wednesday 7 p.m. If the Saturday 11 a.m. hands-on Italian class has averaged 20 booked seats, 20 divided by a cap of 10 gives two chef-instructors. If the Wednesday evening knife-skills class averages 8, the cap of 8 for advanced work gives exactly one instructor. Round up, always — 21 students at a cap of 10 is three instructors, not two-and-a-bit, unless you are willing to close registration at 20.

Step three: layer support staff by trigger, not by habit. Support headcount follows rules, not feel. Add one prep-and-dish assistant to any block running two or more instructors — that person portions mise en place ahead of the class, runs the dishwasher during, and resets stations between courses so the chefs never stop teaching. Add a front-desk or check-in person for any block over roughly 16 students or any block where retail, waivers, or a bar service runs alongside. Add a second dish position when the class serves a full plated meal rather than tastes, because plated service roughly doubles the ware volume. A single class of 20 hands-on students, then, resolves to two instructors, one assistant, one front-desk — four people, not six.

Step four: place those bodies against the real workload curve. The count tells you how many; the kitchen rhythm tells you when they clock in and out. Prep front-loads: someone portions, labels, and preheats 60 to 90 minutes before a hands-on class starts. Instruction is the block itself. Cleanup back-loads hard — every station, pan, and dish resets in 30 to 45 minutes after the last student leaves. That means your assistant's shift starts before the instructors' and may end after them, and your instructors do not all need to be there for the full envelope. On a back-to-back weekend day with three classes, you staff the assistant straight through the day rather than three separate shifts, because the changeover between blocks is the busiest 20 minutes in the building.

Run that loop across every block on the calendar and you have a full week without a single judgment call left in it. The whole exercise takes about an hour the first time and fifteen minutes a week afterward.

Labor cost, hours, and the ranges that tell you it's working

The staffing count is only half the question. The other half is what those bodies cost relative to what the class brings in, and whether the shift lengths you are paying for match the work that actually happens.

Start with shift length. A hands-on evening class that runs two and a half hours does not produce a two-and-a-half-hour shift. The realistic envelope is prep 60 to 90 minutes ahead, the class itself, and 30 to 45 minutes of reset — so a 2.5-hour class generates roughly a 4.5- to 5-hour shift for whoever opens and closes it, and something closer to a 3-hour shift for an instructor who arrives 20 minutes before doors. Scheduling everyone for the full envelope is one of the most common and most expensive mistakes in this business. If you have two instructors and one assistant on a block, the assistant carries the full 5 hours, one instructor carries 3.5, and the second instructor carries 3. That's 11.5 paid hours, not 15.

Now put a cost against it. Instructor pay varies enormously by market and by whether the person is a working chef teaching on the side or a full-time staff educator, and you should price against your own local market rather than a national figure. What travels across markets is the *structure*: instructors cost meaningfully more per hour than prep-and-dish support, which is why the assistant trigger matters. Moving a task from an instructor's hands to an assistant's hands — dishwashing, station reset, portioning — is the single highest-leverage cost move available to a cooking school, because you are paying the cheaper hour to do the work that does not require the expensive skill.

Track the result as a percentage. Take the direct labor cost of a class block — every wage hour attributable to it, loaded with payroll taxes — and divide by the revenue that block generated. Food service operations commonly watch labor in the high twenties to mid thirties as a percentage of revenue, and a teaching kitchen with a healthy per-seat price should be able to sit comfortably inside that band because your revenue per guest is high relative to a restaurant cover. If a specific class type consistently lands above the mid-thirties, you have one of three problems: the class is underpriced, it is under-enrolled, or you are over-staffing it. Look at the enrollment number first, because a 12-seat class staffed for 20 is an enrollment problem wearing a staffing costume.

Watch a second number alongside it: labor hours per student. Divide total paid hours for a block by the students who attended. A well-run 20-student hands-on class at 11.5 paid hours gives you roughly 0.58 hours per student. The same class run at 8 students with the same crew gives 1.44 hours per student — nearly triple. That ratio is the earliest, cleanest warning that a slot should be consolidated, moved, or cut. It catches problems that the percentage alone hides, because it does not move when you raise prices.

Overtime deserves a separate watch. Cooking schools run heavy on weekends, and a part-time instructor who picks up Friday evening, Saturday morning, Saturday afternoon, and Sunday can cross 40 hours in a week without anyone noticing until payroll. Set an alert at 32 or 35 hours so you can see it coming, and understand that overtime rules, break requirements, and rules governing minors vary by state — some states impose daily overtime thresholds, not just weekly. Whatever scheduling tool you use, turn those guardrails on rather than tracking it in your head.

Finally, budget for the substitution rate. Cooking schools lean on part-time instructors who have other jobs, and a certain percentage of shifts will need coverage on short notice. Keep a bench of two or three cross-trained people who can take a block, and know in advance which class formats each of them can actually teach — the person who can run a pasta class is not automatically the person who can run laminated dough.

Where cooking schools get the staffing wrong

Scheduling to capacity instead of expected attendance. The room seats 16, so two instructors get scheduled every time. But the Wednesday class has averaged 8 for two months. You just paid an instructor to watch another instructor teach. Always divide the *expected* number, never the capacity number.

Never scheduling down. Owners readily add a body when a class sells out and almost never remove one when a class softens. Staffing ratchets upward over a season and nobody notices, because each individual addition was defensible. The quarterly review exists to walk it back.

Treating cleanup as free. Cleanup does not happen by magic between classes, and if you have not scheduled someone for it, it happens on the instructor's unpaid goodwill until that instructor quits. Put the reset window in the schedule as paid time on a specific person's shift. The fix is cheap — it is usually 30 to 45 minutes of the least expensive labor in the building — and skipping it is how you lose good chefs.

Over-staffing the small class instead of consolidating it. A six-student class staffed with an instructor and an assistant is a losing block almost regardless of price. The answer is rarely to cut staff below one instructor — it is to combine the session, move it to a better slot, or cancel it and offer transfers. Set a minimum-viable enrollment for each class type, decide the cancellation cutoff (commonly 48 to 72 hours out, so students have notice), and publish it in your booking terms so cancelling is not a fight.

Letting instructors build their own schedule. Well-meaning instructors schedule friends, over-cover their own classes, and under-cover the blocks nobody likes. The ratio card removes this entirely, and it also removes the awkwardness — the number decided, not the manager.

Ignoring class-format differences. Running one universal ratio across demo, hands-on, advanced, and kids' classes guarantees you are wrong in at least two directions simultaneously: over-staffed on demos, under-staffed on advanced technique. Format-specific caps take ten minutes to set and eliminate the entire class of error.

No mid-shift plan for the no-show. Attendance forecasts miss. When four students no-show a 20-seat class, the second instructor is already in the building. Decide in advance what that person does — shifts to prep for the next block, takes on deep-cleaning, or gets cut with a partial-shift understanding you have communicated up front. Ambiguity here is how you end up paying for four hours of standing around, and in some jurisdictions reporting-time or show-up pay rules apply when you send someone home early, so check what applies where you operate before making cutting a routine practice.

Reviewing staffing annually instead of quarterly. Class mix changes, corporate buyouts come and go, a new format catches on. A ratio card set in January is stale by summer. Quarterly is the right cadence, plus an immediate review whenever you launch a new class format.

Choosing the staffing pattern for a given block

Not every block resolves the same way, and the decision tree below covers the cases that recur. The governing question is always the same: what is the format, how many people are actually coming, and what triggers does that combination fire?

A few cases worth calling out explicitly. Back-to-back weekend blocks should never be staffed as independent shifts for support roles — one assistant across the whole day beats three separate short shifts, because the changeover between blocks is the peak workload and because you avoid paying three separate travel-and-setup ramps. Corporate buyouts and private events run off the contract, not the ratio; those groups typically want more attention, often include alcohol service, and should be priced with the extra body built in rather than absorbed. A brand-new class format gets staffed one level richer than the ratio suggests for its first two or three runs, because you do not yet know the intervention frequency; drop to the ratio once you have watched it run. And the sold-out class with a waitlist is the one case where the correct move might be adding a second session rather than a second instructor — check whether the extra instructor's fully-loaded shift cost is actually covered by the four extra seats before you agree to it.

Whatever pattern you land on, publish the schedule far enough ahead that people can plan — two weeks is a reasonable target for a part-time culinary roster, and predictive-scheduling laws in some cities impose specific notice requirements and premium pay for late changes. Check whether any apply to you.

Related questions

What ratio should I use for a demonstration-only class?

Demonstration formats where students watch and taste can run around 16 students per instructor, sometimes higher for lecture-and-taste formats, because no one is handling knives or hot pans. Sightlines and audio matter more than supervision distance. Add a support person if you are plating tastes for the room.

Do I need a dedicated dishwasher for every class?

No. Trigger it on service format, not headcount. Tasting-portion classes fold ware into the prep-and-dish assistant's duties. A full plated meal roughly doubles ware volume and justifies a second dish position, especially on back-to-back blocks where the reset window is tight.

How far ahead should I publish the schedule?

Two weeks is a workable target for a part-time culinary roster and gives people time to arrange other work. Some cities have predictive-scheduling ordinances that mandate specific notice periods and premium pay for late changes — check what applies in your jurisdiction before setting policy.

What is the minimum enrollment before I cancel a class?

Set it per format based on where the block stops covering its own labor plus food cost. Publish the number and a cutoff — commonly 48 to 72 hours out — in your booking terms, and offer transfers rather than refunds where you can, so cancelling is routine rather than a negotiation.

How do I schedule for a class with a mix of skill levels?

Staff to the harder half. If a third of the room is genuinely beginner in an intermediate class, the intervention frequency rises toward the beginner rate, so use the lower cap or add an assistant. Better long-term fix: tighten class descriptions and prerequisites so mixes are rarer.

FAQ

How many employees should I schedule each shift at my cooking school?

Divide expected attendance by your per-instructor cap for that class format, round up, then apply the support triggers. A 20-student hands-on class at a cap of 10 gives two instructors, plus one prep-and-dish assistant because you crossed two instructors, plus a front-desk person because you crossed roughly 16 students — four people. An 8-student weeknight class is one instructor.

Should I schedule off booked seats or off actual attendance?

Off a trailing four-to-six-week average of actual attendance for that specific slot, not booked seats and not room capacity. No-shows are real and they are slot-specific — a Saturday morning class behaves differently from a Wednesday evening one. Booked seats will consistently over-staff you.

What labor percentage should a cooking school run?

Food service operations commonly track labor in the high twenties to mid thirties as a share of revenue, and a teaching kitchen with a strong per-seat price should sit comfortably in that band. Also watch paid labor hours per student — it catches under-enrolled blocks that a percentage alone can hide when prices are high.

How do I handle prep and cleanup without paying everyone for the full envelope?

Stagger clock-ins. The prep-and-dish assistant carries the full window — roughly 60 to 90 minutes before through 30 to 45 minutes after. Instructors arrive shortly before doors and most leave once the room clears. That turns a 15-hour envelope for a three-person crew into something closer to 11 or 12 paid hours.

Do I need extra staff for kids' classes?

Yes. Supervision load per participant is materially higher and the consequences of an unwatched moment are worse. Most operators land between 6 and 8 participants per adult and keep a second adult in the room regardless of headcount. Check whether your state or insurer imposes specific requirements for programs serving minors.

How often should I revisit my staffing ratios?

Quarterly, plus immediately whenever you add a class format. Ratios drift upward over a season because owners readily add a body and rarely remove one. A quarterly review against labor percentage and hours-per-student catches the drift while it is still one or two shifts wide.

Sources

• https://www.bls.gov/ooh/food-preparation-and-serving/chefs-and-head-cooks.htm
• https://www.bls.gov/iag/tgs/iag72.htm
• https://www.dol.gov/agencies/whd/overtime
• https://www.dol.gov/agencies/whd/youthrules
• https://www.osha.gov/restaurant-industry
• https://www.sba.gov/business-guide/manage-your-business/hire-manage-employees
• https://restaurant.org/research-and-media/research/
• https://www.servsafe.com/
• https://www.ciachef.edu/
• https://hbr.org/2015/11/the-truth-about-scheduling-software

Related on PULSE

• How Many Employees Should I Schedule Each Shift at My Music School?
• How Many Employees Should I Schedule Each Shift at My Language School?
• How Many Employees Should I Schedule Each Shift at My Swim School?
• How Many Employees Should I Schedule Each Shift at My Thrift Store?
• Should I open or buy a Celebree School franchise in 2027?
• Should I open or buy a The Coder School franchise in 2027?