Used my own crews on a spec build, and the loss landed in my service P&L

The labor accounting is real, thatch, but I'd put the bigger dollar on the schedule. 15 months on a 2,350 foot single family with an existing plan and a platted lot is a long time, and internal crews are the usual reason. Your techs had a boss with paying customers, and that boss was you, so every time a service call came in the house lost the day. A sub with a signed scope and a retainage schedule doesn't get pulled, or if he does you get to hold money back.

I ran three houses with an in-house framing and trim crew before I quit doing it. The framing was always fine. Everything after framing turned into a queue behind whatever was more urgent that week, and the queue is what costs you.

Same shape at a much smaller scale. I put my own maintenance guy on four unit turns in a 9 unit building because I was paying him anyway. The turns took six weeks each instead of two and I had two vacant units bleeding while he was doing punch list work. When I finally priced his hours at what a turn crew charges, the "savings" were about 1,400 dollars and the vacancy cost was over 9k. Every unit teaches something, and that one taught me that a salaried person is not free labor, he's already spoken for.

What did the 88k lot actually carry in terms of time? You said platted, but platted and permit-ready are different animals in most jurisdictions, and the difference lives in things like utility connection scheduling and whether the county wanted anything on drainage. I ask because 15 months on the build side sometimes has three or four months of that being a lot problem, and people assign it to construction because that's when the calendar started running.

I'll push back a little. The internal labor thing is an accounting insight and it's a good one, but I don't think it's the loss. 441k against a 388k budget is a 13 percent overrun, which on a first ground-up build is honestly survivable. The 6 extra months is the loss. Carry, taxes, insurance, and the risk that your comp set moves under you while you're standing in a half-finished house. I've had two builds where the exit price at month 9 and the exit price at month 16 were different by more than my whole profit. thatch got lucky that the market held.

Dumb question from the paper side. When the build ran long, did the lender just keep charging you or was there an interest reserve that ran dry and then you were writing monthly checks? I'm trying to understand where that 34k actually came from.

thatch, when you say price your own labor at market, what number do you actually put in the budget line? Your billed rate includes your overhead and your profit, so if you use it you're charging the project for profit you're also collecting as the builder. Do you use full billed rate, or loaded cost plus something? I'm building my first budget spreadsheet and this is exactly the cell I keep staring at.

Honest read, at month 6 with the house dried in and your service business bleeding accounts, I'd have listed the shell to another builder and eaten a small haircut. Thin margins are my whole life and the one thing I've learned is that a project that is destroying a different business you own has already cost more than it can pay you. You held it 9 more months to protect 18k.

Did the service company invoice the project at all, even at cost? Asking because I've got six months of research and zero deals, and the thing I'm noticing is that the people who lose money on paperwork usually lost it because two entities they own were doing business with each other on a handshake. Sounds like the money moved but the invoices didn't.

The sensitivity is what I'd have wanted on paper before the lot closed. Build the budget at 9 months and also at 15, with carry, and also at a 10 percent hard cost overrun, and see what the worst intersection looks like. thatch's worst intersection was roughly break-even before you even count the service business, and that's knowable in advance with a spreadsheet and an afternoon. The internal labor discount just moved the whole grid up one notch and made the bad cases look fine.

Reading this from a market where 628k isn't a house, it's four houses. My whole price band is 90 to 180k and a 53k swing would be the entire project. Doesn't change your lesson, just means the cushion you had to absorb it with doesn't exist everywhere.

Good trades hate being the in-house crew on the boss's side project. There's no scope, no change order, and no end date, and they know the service calls are what actually pays them. I've watched two shops lose a lead tech to a spec house that dragged. Ask your techs what those months were like, thatch, you might find the 71k has a tail on it.

I'd just never combine the two. Boring, works. If I own a service business I want the service business getting all of it, and if I want a house I want to be a customer of a builder who signs a fixed scope.

How would you have known in advance which accounts were at risk? I'm trying to work out where I fit and this thread makes it sound like the service side has this hidden fragility where losing two contracts is what breaks the year.

The line I'm writing down is that if the deal only works because your people work for free, it doesn't work. Thanks thatch.

Different angle, thatch. If you'd taken that same summer and managed someone else's spec build for a fee, with your crews staying on their routes, what would that have been worth? I keep looking at the service side of development as the safer seat and this thread is the first time I've seen the actual comparison sitting in one place.