Almost everything in a commercial kitchen can be swapped. A fryer can be replaced with a bigger fryer. A prep table can be cut down. A walk-in cannot, because the room was built around it, the refrigerant lines were run to it, the slab may have been poured for it, and the panels were manufactured to a plan.
So the walk-in is the piece worth getting right first, and it is the piece most often sized by guessing.
Size from deliveries, not from seats
The per-seat rules of thumb in circulation — a square foot of cooler per seat, a fraction of a square foot per meal — are unreliable because they leave out the variable that dominates: how often you take delivery.
An operator receiving produce three times a week and protein twice needs a fraction of the cold storage of one receiving everything on a single Tuesday drop. Same menu, same covers, materially different box.
The inputs that actually determine the size:
Delivery cadence per category. Produce, protein, dairy, beverage and frozen all arrive on their own schedules. Your box has to hold the largest simultaneous accumulation, which usually means the day after the biggest delivery — not an average week.
Volume per delivery, in cases. Not dollars, not weight. Cases, and their dimensions. A produce-heavy menu is a volume problem; a protein-heavy menu is a weight-per-shelf problem. They size differently.
Whether you prep in-house. Scratch prep means bulk ingredients in and prepped product waiting in labelled containers, often both at once. Buying prepped cuts your cold storage substantially and costs you on food.
What has to be in the walk-in versus the line. Undercounter and prep-table refrigeration takes daily working stock out of the walk-in. That trade — bigger line refrigeration, smaller walk-in — is often the right one, because line refrigeration is replaceable and a walk-in is not.
Your growth case, honestly stated. Not the optimistic one. The walk-in is the piece where the optimistic case is worth paying for, because the downside of a box that is too small is renting a second one or rebuilding the room.
The geometry you cannot cheat
Once you know roughly how much you need to store, the box’s dimensions are set by shelving and human beings, and those numbers are not negotiable.
Shelving is 24 inches deep as standard, occasionally 18 or 21 in tight boxes.
An aisle needs 30 inches absolutely minimum, and 36 to actually work in — enough to open a container, turn with a full sheet pan, or get a hand truck down. Thirty inches is what you accept in a small box; it is not comfortable.
Which gives you the real constraint: a box narrower than about 7 ft 6 in only takes shelving on one side. Two facing runs of shelving plus a workable aisle wants around 8 feet of internal width.
This is why walk-ins come in the sizes they come in. Panels are manufactured in standard widths, so nominal box sizes step in roughly 2-foot increments — 6×8, 8×8, 8×10, 10×12 and up. In practice:
- 6×8 is a storage closet you reach into. Fine for a café.
- 8×10 is the smallest box two people can work in at the same time.
- Anything larger is a question of how much shelving you need, not whether you can move.
The door needs roughly 34 inches clear, which will pass a sheet pan rack, and it needs a route from receiving that does not run through the cook line. A walk-in door that opens into the main working aisle creates a collision every delivery and every service.
A cooler and a freezer are not the same build
They look the same from outside. They are not.
A cooler holds 2 to 4 °C. That is set by food safety, not preference — Ontario’s food premises regulation requires potentially hazardous food to be held at 4 °C or below (or 60 °C or above). Your box has to hold that temperature with the door opening constantly on a July afternoon, which is a design condition, not a nameplate.
A freezer holds around −18 °C, and needs an insulated floor. Always. A freezer set directly on a building slab drives frost into the concrete, which heaves it and leaves you with condensation and eventually a failed floor. This is not a corner anyone cuts twice.
An insulated freezer floor sits above the room floor, which means one of two things:
A ramp. Cheaper, immediate, and a permanent daily nuisance — every delivery and every hotel pan goes up it.
A recessed slab. The concrete is cut and repoured so the box floor finishes flush with the kitchen floor. This is the right answer operationally and it is a construction project: saw-cutting, disposal, forming, pouring, and in an upper-floor or podium unit a structural review before anyone cuts anything. It also has to be decided at drawing stage, because it happens before the box arrives.
A cooler can sometimes sit on the building slab with a properly sealed and coved junction, which avoids the ramp question entirely. Whether that is acceptable depends on the slab, the detail, and your health inspector’s view of it as a cleanable surface. Confirm it rather than assume it.
A combination cooler-freezer shares a partition and sometimes a vestibule. It is cheaper than two separate boxes and it means the freezer door opens into the cooler, which costs you cooler floor space and puts a −18 °C door in a 4 °C room. Sometimes the right trade, often chosen for the wrong reason.
Remote versus self-contained
This is the decision on a walk-in quote that most affects both cost and how the kitchen feels to work in, and it is frequently made by whoever prepared the quote rather than by the operator.
Self-contained puts the condensing unit on the box — top-mount or side-mount, factory-assembled, charged and tested. It is the cheapest option, the fastest to install, and the easiest to service. And it rejects all of its heat into the room the box sits in.
That last point is the whole decision. A condensing unit removing heat from inside the box has to put that heat somewhere, and self-contained puts it into your back-of-house. In a small, enclosed storage room with no cooling, that room climbs, the condenser works harder in its own exhaust, efficiency falls, and eventually you are cooling a room to protect a compressor.
Remote puts the condensing unit on the roof or an exterior pad, with refrigerant lines run to the evaporator inside the box. It costs more — line sets, a rack or a curb, electrical run to the roof, possibly structural work, and a refrigeration contractor commissioning a field-charged system rather than plugging in a factory one. In exchange the heat and the compressor noise leave the building entirely.
The room usually makes this decision, not the budget. A walk-in inside a sealed interior room with no mechanical cooling has to go remote, or you buy the cooling instead and pay for it every month.
Two things remote brings with it: a roof penetration, which means the landlord’s consent and the roof warranty, and a rooftop equipment location, which means it competes for space with your exhaust fan and make-up air unit. That is a coordination question, best asked while the mechanical drawings are still drawings. See commercial kitchen ventilation, explained for what else is already going up there.
The services a walk-in needs
A box does not just need floor space. It needs:
A dedicated electrical circuit sized for the condensing unit, plus the evaporator fans, the light, the door heater on a freezer, and the defrost cycle. Freezer defrost is an intermittent load that surprises people. See power and gas for a commercial kitchen.
A condensate destination. Every refrigeration system produces water and it has to go somewhere legitimate — not onto the floor, and on a freezer, through a heated line so it does not freeze in the drain. A missing condensate route is one of the most common late-stage plumbing surprises.
A temperature monitoring point that someone can read without opening the door, and that generates the log your health inspector will ask for.
Clearance around the condensing unit for airflow and service. A self-contained unit shoved against a wall will fail early.
What people get wrong
Sizing to today’s menu. The walk-in is the wrong place to be precise. Every other refrigeration decision is reversible.
Ordering before the room is final. Panels are cut to a plan. A dimension change after release is a change order and a new lead time.
Forgetting the box has to get in. Panels arrive flat and have to pass through the door, round the corner, up the stairs or into the freight elevator. Measure the route, not just the destination. This has stopped more deliveries than it should.
No plan for shelving. Shelving bought after the box arrives routinely does not fit the aisle, and it should have been part of the same layout.
Treating it as a late order. A walk-in is built to order, and the floor finishes, the coving and often the electrical rough-in depend on it being in place. It is an early long-lead item, and it gates work around it. See equipment lead times are your schedule.
Assuming the previous tenant’s box is usable. Sometimes it is an excellent inheritance. Sometimes it is a twenty-year-old box with a failing condensing unit, panels that no longer seal and a floor that will not pass inspection — and it occupies exactly the space you needed. Get it assessed before you count it as an asset.
The short version
- Size from delivery cadence and case volume. Per-seat rules of thumb ignore the variable that matters.
- About 8 feet of internal width is what two facing shelving runs and a workable aisle need. 8×10 is the smallest box two people can work in.
- A freezer needs an insulated floor, so decide ramp versus recessed slab at drawing stage, not on delivery day.
- Self-contained is cheaper and heats the room it sits in. That is usually what forces the remote decision.
- It is built to order and other trades wait on it. Order early.
For where walk-ins sit in the wider refrigeration package, see refrigeration and walk-in coolers and freezers. For a planning range on a fit-out, use the cost calculator.
