Last week I said the design freeze is the single biggest thing that decides whether a modular project makes money. A few of you wrote back some version of “yes, and it’s also the thing nobody warns us about until it’s too late.”
Fair. Let’s fix that.
Somebody in this industry described it as “the day you put your pencils down,” and I’ve never heard it said better.
It’s a date. Before it, everything is open — layouts, structure, mechanicals, window packages, finishes, the works. After it, the drawings we’re building from are the drawings, and changes stop being decisions and start being change orders.
That’s it. It isn’t a legal maneuver, and it isn’t us being difficult. It’s the moment the project stops being an idea and becomes a bill of materials.
Every habit you’ve built on site-built work runs against it.
On a stick job, the sequence protects you. You break ground while the interior is still being argued about. You can be four months in and still deciding on tile, because the tile decision doesn’t have to happen until somebody’s standing in the room. Deferring is free, or close to it.
Factory work inverts that. Design, engineering, third-party plan approval, procurement, and production all sit upstream of the day the first truck rolls. Nothing on your site is holding a place for a decision you haven’t made yet. The National Institute of Building Sciences found modular scopes typically get locked in three to six months ahead of production, with few, if any, change orders after.
Three to six months. That’s the trade. You are buying schedule certainty with decision-making that happens earlier than you’re used to.
Most developers I talk to understand that in principle and still underestimate it in practice. Not because they’re careless — because on every other job they’ve ever run, late was normal and survivable.
This is the part I wish more people could watch happen, because it’s not one cost. It’s four, and they stack.
Engineering redraws it. Days, sometimes a week. Fine on its own.
The plans go back for third-party approval. Our buildings are inspected and approved to your state’s code before they ever leave the plant. A changed drawing is a changed approval. That clock isn’t ours to control.
Materials have their own lead time. If the change touches a window package, a truss, a panel —anything ordered long—you now wait on a supplier who doesn’t care about your closing date.
And you lose your slot. This is the one that hurts, and the one almost nobody sees coming.
Our line is booked. Your modules have a date on it, and behind you is somebody else’s project with their own crane, their own set crew, their own lender. When a change knocks you off your date, you don’t get delayed by the length of the change. You get delayed to the next opening.
That’s how a two-week window revision turns into a two-month schedule hit and a winter set you didn’t plan on. The change was small. The consequence wasn’t.
Here’s the part I’d ask you to sit with, because it connects to last week’s post.
We can hold a hard number on a third to a half of your project cost because the scope is hard. That’s the whole mechanism. We buy the materials, we book the labor, we take the escalation risk — and we can do that precisely because we know exactly what we’re building and when.
An open scope can’t be fixed-priced by anybody. Not by us, not by your GC (General Contractor), not by anyone who intends to still be in business next year.
So the freeze isn’t rigidity we’re imposing on you. It’s the thing you’re trading for certainty. From my chair, that’s a good trade, but it’s only a good trade if you go into it knowing that’s what it is.
The projects that hit their freeze date do a handful of things differently. None of them are complicated:
That last one is the trick most people miss, so I’ll say it plainly.
You can’t defer everything, but you can choose where the undecided things live. Anything that gets installed on site after the set — appliances, some fixtures, flooring in certain configurations, exterior finishes — can stay open a lot longer, because changing it doesn’t touch a drawing or a production slot.
So do that on purpose. Sit down before the freeze and sort the open items into “must be decided now” and “can be decided later, on site.” Move as much as you reasonably can into the second bucket, and then genuinely close the first one.
That’s not gaming the freeze. That’s designing around it, and it’s what the teams who do this repeatedly all figured out eventually.
The design freeze is not the constraint on your modular project. It’s the price of everything modular gives you — the fixed number, the compressed schedule, the crew that shows up in February.
Pay it early and on purpose, and the rest of it works the way it’s supposed to.
Next week I’m going to get into the HUD rule I mentioned in the first post — the proposed change to what counts as a “manufactured home,” why it matters to anybody building to state code, and where I come out on it. That one’s going to be less friendly.
As always, if I’ve got something wrong, tell me.
— Thatcher
