Spec build underwriting that starts and ends with a price-per-square-foot number pulled from your last three projects is underwriting on stale data. Adam Jaspon of Centerline Development, who builds in and around Boston, put it plainly: a project he would have priced around $300 a square foot may now come in well over $350. Nothing about the house changed. Material costs, labor costs and permit fees did.
The fix is not a bigger contingency. It is moving real pricing earlier in the process — rough numbers from your most trusted subs off a schematic set, before civil, structural and MEP drawings exist — and then running a cost-coded budget detailed enough to catch the line items a blended square-foot figure buries.
Below: the pricing sequence Centerline uses to answer “is this even possible,” the budget detail that separates a real feasibility test from a guess, where the Boston-area spec margin currently sits, why zoning is the variable most developers underestimate, and how client work changed the cash flow picture.
Key takeaways
- Historical price per square foot is a lagging indicator. Jaspon’s own benchmark moved from roughly $300/sf to well over $350/sf on comparable work, driven by materials, labor, and permit fees in affluent towns and inside Boston proper.
- Before civil, structural and MEP drawings exist, get rough pricing off a schematic set from your site contractor, framer, plumber, electrician and HVAC sub — that replaces $/SF as the go/no-go test. Hard pricing follows once detailed plans are done.
- Cost-code the budget line by line: pre-rocking for fire rating, one-hour doors versus standard interior doors. A general number against "blueboard" or "doors" hides the small misses that become live cash flow problems.
- In Boston, plan on a denial. Budget 45 to 90 days, file the appeal immediately, and contact the neighborhood liaison the moment you file — competitors skip the market because they don’t know the sequence.
- Client construction management and custom home work now make up about half of Centerline’s construction volume, which put the partners on payroll and let them self-fund more spec equity instead of raising close to 100%.
From the Real Estate Pros Show
This article draws on an interview with Adam Jaspon of Centerline Development and Services on the Real Estate Pros Show, hosted by Scott Bursey.
Why Historical Price Per Square Foot Stopped Working
Initial feasibility at Centerline was always derived from a price per square foot taken off previous projects. That method has broken down. “Deals that we would typically analyze at a rough price per square foot aren’t penciling out to that price per square foot,” Jaspon said. Once the budget gets assembled, the numbers land significantly higher.
His own benchmark shifted: work he was building at roughly $300 a square foot may now run well over $350. That is a 17% swing on the largest cost input in the model, and it does not show up in the historical figure until after you have already bought the dirt.
Three drivers, in his read of the Greater Boston market:
- Material costs, which he describes as consistently climbing rather than spiking and settling.
- Labor costs, compounded by availability — some subs have struggled to keep crews on their teams.
- Permit and municipal fees, heaviest in affluent suburbs and inside Boston proper, which he calls extremely difficult to build in.
The failure mode matters more than the percentage. A stale $/SF number does not make a bad deal look bad. It makes a bad deal look feasible on the front end, so you tie up the site, pay for design, and only discover the gap when the first real budget comes together. By then your options are a thinner margin, a value-engineered product that misses the comp set, or a walk that costs you the deposit and the design spend.
Pulling High-Level Pricing Off a Schematic Set
The replacement for the $/SF go/no-go step is a phone call to five subs while the drawings are still schematic. Centerline gets into a detailed budget earlier in the process specifically to determine whether a project pencils, and that means pricing before the design package is anywhere near complete.
The sequence:
- Architect produces a schematic set. No civil, no structural, no MEP.
- Send it to the most trusted and reliable subs only — site contractor, framer, plumber, electrician, HVAC. These are the five trades that carry the most cost risk and the most site-condition risk.
- Ask for a rough idea, not a bid. The question is “is this project even possible,” not “what is your number.”
- Once civil, structural and MEP drawings are complete, go back to the same subs for hard pricing.
Two things make this work. First, the subs have to be people you have a real relationship with — Jaspon has worked with many of his for five, six, seven years. Nobody prices a schematic set for a builder who bids every job out to three strangers. Second, you have to accept a rough answer. A framer looking at a schematic can tell you whether you are in the neighborhood or off by 20%, and off by 20% is all you need to know at that stage.
The practical benefit is that you learn the deal is dead before you pay for a full design package, and you learn it from the people who will actually do the work rather than from a spreadsheet built on 2023 pricing.
Deals that we would typically analyze at a rough price per square foot aren’t penciling out to that price per square foot. Where we may have been building at 300 a square foot, that project may be well over 350 a square foot now.
— Adam Jaspon, Centerline Development
Cost-Coded Budgets: The Line Items $/SF Hides
Centerline’s budgets started very basic, without much breakdown. They now carry cost codes for every item, and that shift is what turns a pre-construction budget into something you can underwrite against.
Jaspon’s examples are the kind of thing a blended square-foot figure swallows whole. In a multifamily building, you need to account for pre-rocking to achieve a fire rating. You need to specify one-hour doors rather than a typical interior door. Neither shows up if your budget line reads “blueboard” with a general number against it, or “doors” with another. Both cost real money.
His marine engineering background is the mechanism here, not a credential. Eight years offshore meant he could read plans, and he understood how plumbing, HVAC and electrical systems function before he knew residential code. “Nobody knows all of the code,” he said, “but if you have a basic understanding of residential code, you’ll know all the little items to include in a budget.” The first pass on a Centerline budget is broken out that way because they know from the plans what the building actually requires.
The consequence of getting this wrong is not an abstract margin problem. Jaspon named cash flow as the thing that keeps him up at night, and one of the fixes he is working on is improving budgets to ensure every item is captured — because they still occasionally discover they missed little things, and those add up. On a spec deal where you get paid once, at sale, 18 to 20 months out, missed line items are money you carry.
Where the Boston-Area Spec Margin Actually Is
Centerline pivoted away from the density model and away from condos toward luxury single-family. The reasoning is straightforward: the condo market was very strong and has softened, while single-family strengthened significantly post-COVID. The opportunity band Jaspon points to is roughly $1.5M to $4M.
The model is a teardown of obsolete suburban stock. Take an older three-bed, one-bath home in a Boston suburb, raze it, and build a four-bed or five-bed, four-and-a-half-bath house sized for how families actually live now — home office, home gym, open kitchen-living. Centerline also spends deliberately on garage space, including epoxy floors and heated garages.
The value creation is the gap between what the existing house does and what the lot could support. An obsolete floor plan on a good street in a strong town is a design problem, not a market problem, and Jaspon notes that once they tried new construction they rarely bothered with renovation again on the development side.
Rentals run on the opposite logic. New construction is very hard to pencil on rentals in the Boston area, so Centerline renovates instead — and there, density is still the play. Demand has shifted: luxury rentals have definitely softened, while there is very high demand for affordable apartments in what he calls the class B and C market, with people moving further outside the city. Their 22 doors sit across four properties, two in Dorchester and two about 90 minutes out.
Same operator, same market, two directly opposite build-versus-renovate answers depending on exit.
Zoning as the Real Underwriting Variable
Asked what most developers miss on acquisition, Jaspon did not name a design element. He named zoning. “Familiarizing yourself with the zoning process in whatever market you’re playing in is really the secret.” Someone unfamiliar with a market cannot interpret the code well enough to see the hidden value.
Boston is his example because the process is knowable and most people don’t bother. His sequence:
- Expect the denial. You will submit and get denied. Underwrite on that basis from the jump.
- Budget 45 to 90 days for that stage.
- File the appeal immediately on receiving the denial. No dead time.
- Reach out to the neighborhood liaison as soon as the appeal is filed, and build those relationships early so the process runs seamlessly.
The second half is knowing when to handle zoning in-house and when to bring in help. That means knowing who the zoning attorney is in that specific neighborhood or community who gets it done, and having a civil engineer who knows how to navigate the same process. Those are community-level relationships, not statewide ones.
Here is the arbitrage: a lot of builders stay out of Boston entirely because they don’t understand the process. The denial reads as a dead deal to someone who hasn’t done it before, so they never bid. If you have priced the 45-to-90-day delay into your carry and you know the appeal path, you are competing against a thinner field for the same site. Jaspon also credits Build Boston, a local group of builders, developers, architects and engineers, as a source of permitting and zoning advice when he hits something new.
Capital, Cash Flow and One Expensive Compliance Lesson
Capital is available; it costs more. Centerline contributes its own equity and raises the balance on larger deals, and Jaspon is direct about the repricing: deals they used to raise money at 12% for aren’t priced there anymore. Cost of capital, not access to capital, is the constraint.
The more useful structural change was adding client work. Centerline began taking on construction management for other developers and custom home building for residential clients about two years ago, and it now makes up roughly half of construction volume. Development-only income arrives when a deal sells — every 18 to 20 months. Client work arrives continuously.
Two consequences followed. The partners put themselves on payroll and can project income. And because client revenue funds operations, they now self-fund a much larger share of their spec equity: “Where we used to raise almost 100% of the capital needed, now we don’t need to do that anymore.” Less raised equity at higher rates means more of the spec margin stays in the deal.
His most expensive lesson is a compliance one. Centerline raised money from friends and family and did not file the correct paperwork with the state; the state determined the forms were wrong and they were fined. He describes it as expensive and as a lesson learned — they now document and report everything carefully.
That is his experience, not legal advice, and securities filing requirements vary by state and by how a raise is structured. If you are taking money from anyone, including people you know well, run the structure past a securities attorney before the first wire.
Frequently asked questions
How early should you get real subcontractor pricing on a spec build?
Before the design package is finished. Centerline pulls rough pricing off a schematic set from the architect — no civil, structural or MEP drawings yet — from its site contractor, framer, plumber, electrician and HVAC sub. The purpose is answering whether the project is possible at all, not producing a bid.
Hard pricing comes later, from the same subs, once detailed plans are complete. Doing it in that order means you find out a deal is dead before you have paid for a full set of drawings.
What should a pre-construction budget include that a price-per-square-foot estimate misses?
The code-driven and spec-driven items that a blended figure averages away. Jaspon’s examples from multifamily work: pre-rocking to achieve a required fire rating, and one-hour rated doors instead of typical interior doors. A budget that carries a single general number for “blueboard” or “doors” prices neither.
The fix is cost codes for every item, built from actually reading the plans. Small misses are the ones that show up as cash flow problems later, since a spec deal pays out once, at sale.
How long does the Boston zoning appeal process take?
Jaspon budgets roughly 45 to 90 days for the submit-and-deny stage in Boston, and underwrites on the assumption the initial submission will be denied. The appeal should be filed immediately on receipt of the denial, followed right away by outreach to the neighborhood liaison.
Timelines vary by community and by what you are asking for, so treat this as one operator’s working assumption for Boston rather than a rule. Whether you need a local zoning attorney and civil engineer depends on the specific neighborhood.
Is it better to bid every trade out or stay with the same subs?
Jaspon does bid items out, but he is explicit that he does not require three prices for everything: “Sometimes we know where pricing should be and we just need to move with the people we trust.” Many of his subs have been with him five to seven years.
The trade-off he accepts is paying more for reliability. His plumber probably isn’t the cheapest, but he shows up when called and resolves homeowner issues immediately. That relationship is also what makes early schematic-stage pricing possible — a sub you rebid every job won’t price a rough set for you.
Why is new construction harder to pencil on rentals than on for-sale homes?
Because the rent has to service the construction cost, and in the Boston area those costs have outrun what the achievable rents support. Jaspon says new construction rentals are very hard to pencil there, so Centerline renovates existing buildings for its rental portfolio while pursuing new construction on the for-sale side.
Demand composition compounds it. Luxury rentals have softened while the strong demand sits in class B and C workforce housing — a segment where new-construction cost basis is especially hard to justify against attainable rents.
The bottom line
Before you underwrite another spec deal, pull the $/SF figure you have been using and date it. If it came off a project that closed more than a year ago, it is a starting hypothesis and nothing more — replace it with rough numbers from five trusted subs off the schematic set, and make that call your go/no-go instead.
