What contingency and carry am I supposed to be putting in a 40-lot pro forma?

On my much smaller version of this exercise the line I kept underestimating wasn't contingency, it was the stuff nobody puts on the estimate at all. Geotech, wetland delineation, traffic study, boundary and topo survey, the plan review fees, the SWPPP inspections every two weeks for the whole build, and the bond premium. Those are all soft and they all show up as checks before you have revenue.

Total beginner question so ignore me if it's dumb, but what's the difference between the fees you quoted at $19k a lot and the improvement cost? I thought the improvements were what you paid the city for.

Taking them in order. On an engineer's opinion of probable cost prepared before you have bids, 10 to 15 percent contingency is the range most lenders will accept, and anything at 5 percent means you have hard bids or you're fooling yourself. Rock is the specific reason. If your geotech shows shallow refusal or a high water table, trench excavation prices move enough to eat a 5 percent line by itself, so the number should be set after you read the borings rather than before.

Carry depends almost entirely on where you are in the approval process. With a recorded final plat and approved construction plans in hand, you're looking at mobilization plus a construction season, and dry utility coordination usually sets the critical path because the power company works on its own calendar. Starting from a rezoning or a preliminary plat, add whatever the local hearing cycle is, and in many jurisdictions that's a year or more with no guarantee at the end. Ask three engineers who work in that county how long the last comparable plat took from application to recorded, and use the slowest answer.

Soft costs at 10 to 12 percent of hard is roughly the right order of magnitude for a straightforward greenfield plat, and it climbs fast if you need environmental work, a traffic study, or offsite easements from neighbors who now know why you're calling.

One thing your numbers don't show at all: $4.72M of revenue only exists if 40 lots sell at $118k, and lot absorption is set by one or two builders in that submarket. Model the deal at 12 lots a year and at $105k a lot, and if it only works at full price and full speed, what you have is a bet on a single buyer's appetite rather than a development budget.

Scope creep on renovation is the same disease as change orders on site work, and the mechanism is identical. Somebody finds a condition, the drawing doesn't cover it, and the sub prices the fix without competition. On dirt the classic ones are unsuitable soils you have to haul off and import structural fill for, and a storm outfall that turns out to need an easement across the neighbor. Both of those are 10 percent contingency events, not 5.

@windlass your $19k of fees, is that a fixed schedule or does the city index it annually? Ours adjusts every July and the increase applies at permit issuance, not at approval, so a slow project pays the new number.

Following this one because I understood maybe a third of it. What's SWPPP?

@rigging stormwater pollution prevention plan. It's the erosion control plan for the site, silt fence and inlet protection and so on, plus the inspection reports you have to keep during construction. There's a whole cottage industry of consultants who do the inspections, which is part of why I keep looking at the service side of this business rather than the developer side.

@windlass the answer to your third question in practice is that soft costs are whatever they are and you find out. What I'd do before touching the model again is spend the money on a geotech and a preliminary layout from an engineer who has platted in that jurisdiction. Two or three thousand for borings answers your contingency question with data instead of a percentage, and the layout tells you whether 40 lots is even the real yield once detention and the road takes its bite. I've seen lot counts come off a napkin sketch by 15 percent after the pond gets sized properly.