Your Floor Plan Is Lying to You: Design the Facility Around the Equipment List, Not the Architecture

Here is a pattern we see so often it stopped being interesting and started being a diagnosis. An operator wins a license, hires an architect, and produces a beautiful floor plan. Months later the equipment arrives, and the building starts arguing with it. The chiller needs pad space nobody drew. The extraction room's panel is sixty amps short. The dehumidifiers have nowhere to reject heat. The change orders begin, and every one of them costs three to ten times what the same decision would have cost on paper.

The diagnosis is always the same: the floor plan was drawn before the equipment list existed. So here is the position we will defend: architects design shells, and equipment schedules design facilities. If your equipment list does not exist yet, your floor plan is a guess wearing a stamp.

Exhibit A: your utilities are equipment-derived

Ask what actually determines a cannabis facility's mechanical and electrical design. Not the square footage. Not the room names. The equipment. Chiller tonnage falls out of the extraction and distillation equipment's heat load. Panel schedules fall out of the motor list. Dehumidification capacity falls out of the irrigation plan and canopy. The classified electrical zone falls out of which solvent the extraction machine uses, a decision that alone swings the electrical budget by a factor of two, as the requirements in our C1D1 requirements guide make plain.

None of those numbers can be derived from architecture. All of them can be derived from an equipment schedule. We made this argument narrowly a while ago in chiller-first facility design, where the case was that process cooling should be designed before the rooms it serves. The general version is stronger: every utility in the building is downstream of a machine, and a design process that starts with walls is deriving the knowable from the unknown. Engineers know this, which is why MEP teams handed a floor plan without an equipment schedule respond by padding everything, and padding is just a change order you prepay without getting the work.

Exhibit B: the change-order math

The economics are lopsided enough that they should end the argument on their own. Moving a wall on a drawing costs an hour of design time. Adding a 100 amp circuit during rough-in costs a line item. The same circuit after drywall costs demolition, patching, re-inspection, and schedule, and construction cost references like RSMeans put the multiplier on rework at several times the planned-work cost before you count the delay itself. In a facility whose flower rooms are waiting on a mechanical retrofit, the delay is the expensive part: a month of slipped harvest in a mid size cultivation runs into six figures of revenue, which is why the industry's own buildout retrospectives, MJBiz's factbook interviews among them, read like a catalog of climate systems corrected after opening. The pattern connects directly to the yield economics we walk through in facility optimization ROI: the buildings that hit their numbers are the ones that never had to stop and fix themselves.

Exhibit C: the used-equipment corollary

There is a purchasing argument hiding inside the design argument. If the equipment list is locked before the drawings are, you can buy used, because you have time to wait for the right machine and design around its known, documented specs. Used process equipment trades at 40 to 60 percent off list, the arithmetic we walk through in our used equipment buying guide, and the savings on three or four major machines routinely covers an entire room's buildout. Design-then-buy reverses this. When the drawings dictate a delivery date and a footprint before the machine is chosen, you buy new, at list, on lead time, because only new equipment can be summoned to spec on schedule. The floor-plan-first process does not just cost change orders; it quietly forecloses the cheapest capital strategy available to an operator.

The counterargument, and where it is right

The standard objection is that sites move faster than equipment decisions: the lease has an option deadline, the license has a buildout clock, and you cannot stall the architect while you shop for centrifuges. Fair, and site-first sequencing is genuinely right about one thing: a building's bones, its power service, clear height, and zoning, can disqualify it no matter what equipment you intend, so the site diligence has to run early and hard.

But notice what that concedes. The site questions that matter, can the service carry the load, can the roof carry the units, can the slab drain, are equipment questions asked at building scale. You cannot even do site diligence well without a draft equipment schedule, because you do not know what load the service has to carry. The honest version of site-first is equipment-informed site selection, which is our position with the steps reordered. What we are actually arguing against is the common case: a floor plan finalized while the equipment list is still a category budget, and a team discovering its machines like plot twists.

The sequence that works

License first, because it bounds what you may build. Product mix second, because it decides what you will actually produce, and every machine follows from a product. Equipment schedule third: every major machine with its power, water, drain, heat rejection, and footprint, drafted even if the final brand is undecided, refined as quotes land. Utilities fourth, derived from the schedule with real numbers instead of allowances. Floor plan fifth, drawn around a process that is now fully specified, the way our room-by-room facility design guide lays out step by step. The floor plan comes last not because it matters least but because it is the summary of every decision above it, and summaries written before the content are fiction.

Operators sometimes hear this sequence as slower. It is not. The equipment schedule takes weeks, runs parallel to license and site work, and repays itself the first time an engineer sizes a system from data instead of allowances. The floor plan drawn sixth is drawn once. The floor plan drawn first is drawn twice and built one and a half times, and the half is the expensive part.

If you are staring at a beautiful floor plan and an equipment budget that is still one line, that is the moment to stop. Bring us the license and the product plan, and we will build the equipment schedule that tells you whether the drawing is a plan or a lie.