On the horizontal work, do you hire one turnkey site contractor or contract the subs yourself

I've done it both ways on small stuff and the number that decided it for me was the coordination premium. Turnkey bid on a 9 lot job came in about 14% over the sum of the individual scopes I'd priced myself. On a deal where my whole margin was around 18% of revenue that's not a rounding error.

I also ate two weeks of standby because the paver showed up to subgrade the utility crew hadn't finished proof rolling, so call it a wash on that job. Second time I built a two week buffer between scopes into the schedule and it went fine.

What does proof rolling mean? I've read this thread twice and I'm following most of it but not that.

@megan_doyle15 it's driving a loaded truck over the compacted subgrade so the inspector can see whether it deflects. If it pumps or ruts, that area gets dug out and replaced before pavement goes on. It's the checkpoint where the earthwork scope either passes or becomes a problem, which is exactly why @nicole_adeyemi's question about who owns the gap between scopes matters.

The thing that decides this isn't the markup, it's whether the buyer can actually schedule. Without someone on that site most days, self-managing multi-trade work doesn't work, and the price difference disappears the first time a crew mobilizes and can't start. With a day job elsewhere, buy turnkey and skip the math on the premium.

Is there a middle option where you hire a construction manager for a fee instead of a GC with markup? I see that on vertical work all the time and I don't know if it exists in land.

@nataliehsu yes, and on horizontal work it's more common than people think, usually a percentage of hard cost or a flat monthly fee. That structure buys coordination without the trade markup, but the contracts stay in the developer's own name, which also means the risk stays there. That's the trade. One thing worth adding: the bond matters here. If the county holds a subdivision improvement bond or letter of credit on the improvements, the surety cares who's building them. A single site contractor with a track record in that county is easier to bond around than a first-time developer coordinating three separate crews. That has priced deals more decisively than the 14% @joe_delgado5 is talking about.

@gregpowell66 that's the point I came here to make. On renovation the scope creep comes from the walls. On horizontal it comes from what's under the dirt, and the difference is that unsuitable soil is measured in cubic yards at a unit price, so the overrun scales fast.

If you self-manage, get unit prices for undercut and import in every earthwork contract before anybody mobilizes. Lump sum with no unit prices is where you find out what a bad geotech assumption costs.

My honest fear, and I have zero deals done, is that I wouldn't know if a turnkey bid was 14% high or 40% high. Is there any way to sanity check a horizontal bid without having done one?

@jordan_ellis2 ask the civil engineer for an opinion of probable cost before bidding it out. They do those routinely and they've seen recent bids in the same county. It won't be exact and it gives a range to test the bid against. If a bid comes back well outside the engineer's range in either direction, that's worth understanding, and low is as worrying as high.

So the engineer both designs the improvements and estimates what they'll cost? Is there a conflict there, or is that just how it's done?

@kevin_lombard it's how it's done, and the design does drive the cost, so a design that's conservative on pipe sizing or road section costs real money. That's an argument for having someone review the plans against the actual county standard rather than against the engineer's habit. A lender's third party review catching an over-designed storm system and taking a meaningful number out of the budget is a common outcome of that kind of check. On the main question, a lender looking at a development loan treats a borrower coordinating three separate trades as a different credit than a borrower with one contract and a schedule. Doesn't mean no, it means more questions and usually more equity.

Variance matters more than the mean here. Self-managed scopes have a lower expected cost and a much fatter tail. On a 22 lot deal where the improvement budget is a third of total cost, a 14% higher number for a tighter distribution is usually worth taking, because the downside case is what kills the deal, not the base case. The answer flips if the developer is genuinely a builder themselves. @joe_delgado5 self-managing is a different bet than @jordan_ellis2 self-managing, and a poll can't capture that.

@dpatel_re that's the clearest version of it. The question isn't which method is better, it's who's holding the schedule.

From the contractor's side, here's the part that rarely gets said out loud. When a developer self-manages and calls for one scope, a contractor prices the coordination risk into that number anyway, since there's no telling who's ahead in the sequence or whether they'll be done on time. So some of that 14 percent premium shows up in the individual bids too. The savings are real on small jobs where the whole site is visible from the road. They shrink fast once the job is big enough that it isn't.