A good cost code structure has two parts: a short list of work codes (what was built) and a fixed set of cost types (labor, material, subcontract, equipment, other). Number them consistently, use them in the estimate and the books, and keep the list short enough that the field codes accurately.
Review invoices, committed costs, and cash timing together before making project decisions. AI-generated illustration.
A construction cost code structure is the list of categories you use to budget and track every dollar on a job. Build it in two layers. The first layer is the work: phases or divisions such as sitework, concrete, framing, or electrical rough-in. The second layer is the cost type: labor, material, subcontract, equipment, and other. A cost lands on one work code and one cost type, for example 06-100 Rough Framing, Labor. Number the codes the same way on every job, use the exact same list in your estimates and your accounting, and keep it short enough that a foreman can pick the right code on a phone in ten seconds.
The structure matters more than the software. Job costing, WIP reports, and estimating history all depend on it. A messy list produces messy numbers no matter what system you run.
Start with a standard or build your own?
Many general contractors base their codes on MasterFormat, the specification numbering system published by the Construction Specifications Institute, which CSI describes as the industry's standardized language for project documentation. Its divisions, such as Division 01 General Requirements, Division 09 Finishes, and Division 26 Electrical, line up with how drawings and specs are organized, which makes buyout and subcontract scopes easier to match. CSI revises the standard from time to time, so check which edition your architects are using.
Trade contractors usually do better with codes built around how their own work happens. An electrician thinks in underground, rough-in, trim, and gear. A residential builder thinks in phases: foundation, framing, dry-in, mechanical rough, drywall, finishes. Either approach works. What matters is that the code matches how the work is estimated and how the crew actually builds it.
How to build your cost code structure in seven steps
1. Pull your last ten estimates
List the line items you actually price. If your estimates have 18 recurring sections, that is your starting point. Codes that never appear in an estimate will never have a budget to compare against.
2. Group them into 15 to 40 work codes
Combine items that you would never make different decisions about. Interior and exterior paint might be one code for a GC who subs it out, and two codes for a painting contractor. Ask: if this line ran over budget, would I want to know separately? If not, merge it.
3. Add cost types as a separate layer
Do not create "Framing Labor" and "Framing Material" as two codes. Create one Framing code and attach cost types to it. This keeps the list short and lets you report labor across all codes in one view, which is where most fade starts.
4. Pick a numbering scheme and leave gaps
Use a pattern like division-section (06-100, 06-200) or phase numbers in tens or hundreds. Gaps let you add a code later without renumbering. Keep the number format identical across estimating, project management, and accounting.
5. Add standard codes every job needs
General conditions: supervision, temporary facilities, cleanup, dumpsters.
Change orders: a code or code suffix so approved change work can be tracked against its own budget.
Punch and warranty: so closeout cost is visible instead of buried in the last trade code.
Rework: optional, but powerful if you want to measure it.
6. Write a one-page code guide
For each code, one line on what goes there and what does not. "06-100 Rough Framing: walls, floors, roof framing, blocking. Not: trim carpentry (06-200), doors (08-100)." Print it, pin it in every truck, and load it into the timesheet app.
7. Lock it, then review once a year
Only one person can add codes. Changes happen at an annual review, not mid-job. Codes that collected almost no cost all year get merged; codes that hold a huge share of cost with frequent overruns may get split.
Worked example: a residential remodeler's code list
Say a remodeler running kitchens, baths, and additions builds a code list. Each code takes the five cost types (L labor, M material, S subcontract, E equipment, O other). This is an illustration, not a standard.
Example remodeling cost codes
Code
Name
Typical cost types used
01-000
General conditions and supervision
L, O
01-100
Permits and fees
O
02-100
Demolition and haul-off
L, E, O
03-100
Foundation and concrete
S, M
06-100
Rough framing
L, M
06-200
Finish carpentry and trim
L, M
07-100
Roofing and exterior envelope
S, M
09-100
Drywall and finish
S
09-300
Tile
S, M
09-900
Paint
S
12-100
Cabinets and countertops
M, S
22-100
Plumbing
S
26-100
Electrical
S
90-000
Change orders
L, M, S, O
95-000
Punch and warranty
L, M
Fifteen codes. The carpenter on site uses four of them on a typical day. The estimator prices the same fifteen. The bookkeeper codes every supplier bill to one of them. After a year, the owner can see that tile ran over budget on 7 of 11 bathrooms, which says more about estimating than any single job report could.
Getting the field to code correctly
The best structure fails if crews guess. Put the codes for each job on the timesheet app or paper sheet, filtered to the codes that job actually budgets, so a framer on a bathroom job is not scrolling past sitework. Have the foreman approve hours daily, not weekly. And close the loop: when the PM sees hours on the wrong code, tell the crew lead that day. Within a few weeks, accuracy improves because people see the numbers are being used.
Mistakes to avoid
Too many codes. A 400-line list guarantees miscoding. Detail you cannot trust is worse than less detail you can.
Codes that differ between estimating and accounting. Then no budget-to-actual comparison is possible without manual mapping.
Mixing work and cost type in one list. You lose the ability to see all labor at once.
A catch-all "miscellaneous" code. It becomes the biggest code on every job. If you must have one, review it weekly.
Changing codes mid-year. Comparisons across jobs and years break.
No code guide. Two people code the same cost two different ways, and both are sure they are right.
Smart Business Automator uses one cost code list across estimates, budgets, purchase orders, timesheets, and job cost reports, so the handoff from bid to job does not require re-keying. See estimate to project handoff automation.
Frequently asked questions
Should I use CSI MasterFormat for cost codes?+
It works well for commercial general contractors because drawings, specs, and subcontract scopes are organized the same way. Trade contractors and residential builders often do better with codes based on their own phases of work.
How many cost codes is too many?+
When the field cannot pick the right code quickly, or when many codes collect almost no cost in a year. Many contractors work with 15 to 40 active work codes plus a short list of cost types.
What are cost types in construction?+
Cost types split each work code by the kind of cost: usually labor, material, subcontract, equipment, and other. They let you see, for example, all labor across a job in one total.
Can I change my cost codes after setting up job costing?+
Yes, but do it between years or between jobs, not in the middle, and map old codes to new ones so historical comparisons still work.
Washington just put $17.5 billion in conditional loans behind a plan to build ten large reactors across five sites - the biggest single push to restart American large-reactor construction in a generation. Who has to put up a billion dollars in equity before a shovel moves, which regions get the work, and the prequalification move every civil, concrete, mechanical, and electrical sub should make before the EPCs lock up the bid list.
Starts muted when the player comes into view.
Read the full episode transcript
Published transcript from Scaling Legends. It may contain transcription errors. View the source transcript.
As always, this content is for educational and informational purposes only. It is not legal, financial, or professional advice. And before we get into it, we're currently accepting beta testers for Smart Business Automator. You can sign up at smartbusinessautomator .com. Beta testers get first access to new features and updates before the public rollout. So you are tuning into a Scaling Legends deep dive. Our topic today, the $17 .5 billion reactor loan program, Westinghouse's five -site bet, and how subcontractors get on the prequel list first. Ten large reactors, five sites, $17 .5 billion in federal loans to make them happen. The biggest restart of American large reactor construction in decades just got its financing, and the contractors who get on the list now are the ones who will be working these jobs for the next decade. It's
just a staggering amount of capital. And, you know, it completely reshapes the landscape for American infrastructure over the next 10 to 15 years. OK, let's unpack this, because when you throw around a number like $17 .5 billion coming out of the Department of Energy, it's really easy to just kind of glaze over. It sounds like monopoly money or, I don't know, some abstract government white paper project that might happen in the year 2050. But this isn't theoretical. I mean, we are talking about shovels in the ground. Oh, definitely. And according to Smart Business Automator's market intelligence report, this is far past the theoretical stage. The core mission of our deep dive today is looking at how you actually deploy 10 Westinghouse AP1000 reactors across five initial sites. And that's two reactors per site, right? Exactly. Two reactors per site.
And we want to look at what that tidal wave of capital means for the businesses listening right now. Because this program is structured around... very specific conditional commitments. The money doesn't just flow blindly from the federal government. Right, which leads to my first question as an operator looking at this space. When I hear about a $17 .5 billion loan facility, I immediately look for the friction. Like, does the government just wire a few billion dollars to a project account and we start pouring concrete tomorrow? Because anyone who has ever done even a municipal road project knows that public money comes with... Very heavy chains attached. Yeah, you're spot on. The government is providing the loan facility, but the actual release of those funds is gated by a massive barrier to entry. So the conditional
commitments in the data require roughly $1 billion in upfront equity per site. Wait, hold on. $1 billion per site? Yeah, $1 billion, before a single federal loan dollar actually moves. Just to turn the tap on, I mean, how does a utility CFO even justify tying up that kind of liquid capital before they have final clearance to move dirt? Well, they justify it because the alternative, which is sequential stop -and -go funding, costs them exponentially more in schedule delays and lost revenue. But the burden doesn't fall entirely on the U2L here. That $1 billion barrier to entry is actually split. Okay, how so? You're looking at roughly $500 million from Westinghouse, the technology provider, and $500 million from the utility partner at that specific site. What's
fascinating here is the sheer level of skin in the game this requires from the vendor. Oh, wow. That forces absolute alignment. Because historically, in mega infrastructure projects, the dynamic is usually... You know, the technology vendor selling their equipment, taking their margin and stepping back. Right. They just wipe their hands of it. Exactly. Then the utility is left holding the bag, trying to get the government to shoulder the deployment risk. If a turbine is delayed or, say, concrete testing fails, there's a lot of finger -pointing and people walking away. And the DOE learned from those historical failures. When both the vendor and the utility have half a billion dollars sitting on the table before the first federal dollar arrives, it changes the psychology entirely. It proves to the Department of Energy and, crucially, to the regional supply
chain that this is a committed partnership. They're pot committed. Yes, exactly. It eliminates the moral hazard of a vendor over -promising on timelines because their own balance sheet is bleeding if the schedule slips. So they are married to each other, for better or worse, because of that equity lockup. Now, looking at the intelligence report, there's a specific operational benefit to structuring it this way. This heavy upfront financing is designed to pull as much as three years off the deployment timelines for these reactors. Which is huge. Three years is an eternity in heavy construction. Yeah. But I want to challenge the mechanics of that. How exactly does having money sitting in an escrow account act as a time machine? Is this like... paying for TSA pre -check for infrastructure. You're paying heavily up front
so you don't spend years standing in the bureaucratic security line. It is like TSA pre -check, but imagine if paying for pre -check also meant the TSA started inspecting your luggage while you were still in the taxi on the way to the airport. Okay, I like that analogy. The time savings don't just come from skipping a line. They come from guaranteed parallel motion. Let's look at why massive infrastructure bills suffer from schedule overruns. It's rarely an engineering failure. It's the start and stop nature of sequential funding. Meaning you build phase one, then you have to pause, put down your tools, and go hat in hand to the regulators or the banks to fund phase two. Right. And when you pause a project of this scale, you can't just hit a freeze button and walk away. Demobilizing a site is an absolute operational nightmare. Tower cranes cost tens
of thousands of dollars a month to rent. You don't leave them sitting idle. No, definitely not. Specialized welding crews don't wait around in a trailer either. They get on a plane and go work on a bridge project two states over. Project management teams are reassigned. And then, you know, six or eight months later, when the next tranche of funding is finally approved, you have to spend millions of dollars and months of time just remobilizing the site. Exactly. You have to rehire, retrain, recertify everybody. And that remobilization friction is where years are lost. By having that billion -dollar equity foundation locked in and the $17 .5 billion loan facility guaranteed behind it, you eliminate the stop -and -go. You move from sequential development to parallel development. You can order the long -lead forging materials for the reactor vessel at the exact same time you are doing the heavy
civil site prep. Because the vendors know the money is already secured. Yes. You don't have to wait for the dirt to be moved before you sign the contract for the steel. But, I mean, parallel development introduces its own kind of friction, doesn't it? If I order the structural steel and the massive steam turbines on day one, and they arrive on month 12... But my site prep was delayed by bad weather. Well, now I have a massive logistics problem. I'm storing incredibly sensitive, multi -million dollar components in a dirt field. That is a very real challenge. And it's actually why lay -down yards and preservation maintenance become massive subcontracts on these sites. You need climate -controlled temporary structures. You even need crews whose only job is to rotate turbine shafts manually every week so they don't warp under their own weight while in storage.
Wait, really? Just manually rotating them. Yeah, it's incredibly intensive. But from a macroeconomic perspective, paying for complex storage logistics is still vastly cheaper than losing three years of power generation revenue because you waited to order the turbine. The capital acts as a shock absorber against timeline friction. Which is going to be a massive signal for the contractors listening to this. Predictable, funded, uninterrupted schedules. But parallel development only works if you have a guaranteed pipeline to keep those parallel tracks fed. Which brings us to the horizon beyond just those first five sites. According to the data, Westinghouse has already signed letters of intent with seven potential partners. Seven additional letters of intent on top of the initial five sites. Here's where it gets
really interesting. Those five initial sites are just the leading edge of a multi -site build -out. We aren't just talking about a few isolated projects in a specific region. We're looking at a national pipeline. And that pipeline forces a total shift in how this work is contracted. Yeah, if we connect this to the bigger picture, we are witnessing a fundamental paradigm shift in how America builds nuclear power. Historically, every nuclear plant in this country was treated as a bespoke project. Custom engineering, unique site layouts, distinct containment structures. You build a one -off masterpiece. Right. But if you're building a one -off, you have zero economies of scale. Exactly. You learn a hard operational lesson on site, maybe a more efficient way to sequence the rebar for the cooling tower. And that lesson is completely useless on site
B because the cooling tower design is totally different. But this entire pipeline is based on the Westinghouse AP1000 reactor. That is a standardized design. So we are moving away from custom artisan construction into an advanced manufacturing mindset. Precisely. The AP1000 is highly modular. It's designed to be built in factories in standardized modules, shipped to the site on rail cars or heavy haul trucks, and then assembled. And because they're deploying the exact same standardized reactor across all these different sites, the physical work of assembly becomes highly repeatable. I want to dig into the compliance side of this, because in the nuclear industry, getting qualified to supply parts or labor is notoriously brutal. The standards are incredibly high. For a subcontractor listening right now, how high is that barrier to entry
to actually work on an AP1000? It is arguably the highest commercial standard in the world. The quality assurance standards, things like NQA1, require an enormous investment. And it's not just about doing good work. It's about proving on paper that the work was done perfectly. Give me a tangible example of what NQA1 looks like on the ground. Because people hear quality assurance and they think of, you know, an inspector with a clipboard checking a box at the end of the day. Sure. Let's take a single... structural bolt holding a pump mount to the floor. Under NQA 1, you don't just buy a bucket of bolts from an industrial supplier. You need a documented paper trail tracing that specific bolt back to the foundry that poured the steel. Wow, back to the foundry. Oh, yeah. You need the material test reports showing the chemical composition of that specific heat of steel. When the bolt
is installed, the worker must use a torque wrench that has a valid traceable calibration certificate. The person reading the torque wrench must have documented documented up -to -date training records. If a single piece of paper in that chain is missing, the bolt must be removed and discarded. That is just an intense level of administrative overhead. I can see why it would take a company a year and hundreds of thousands of dollars just to pass the audits to get on an approved vendor list. And in the old days of bespoke plants, you'd spend a year getting qualified for one specific custom project, you'd do the job, and then that qualification was essentially a dead asset until another custom plant got funded in your region. But look at how the standardized AT1000 rollout changes that equation. Getting qualified for this supply chain means you qualify once. And that qualification carries over to
the next job. And the next site, and the site after that. It is like compounding interest on your compliance investment. The first audit is your principal investment. It hurts. It takes time. It costs money. But every subsequent site is pure yield. If your firm learns how to manage the NQA1 paperwork for assembling the modular components or perfectly pouring the specific nuclear -grade concrete mixes at site number one, Westinghouse, and the lead ETC, the engineering procurement and construction firm, they do not want to go find a new vendor for site number two. Well, they can't afford to find a new vendor. The switching costs would be devastating to their timeline. I mean, why would an EPC risk a six -month schedule delay to train a new concrete supplier on NQA1 documentation when they have a supplier who already knows the exact spec,
already has the safety clearances, and already has the internal QA culture? They wouldn't. The EPC wants the proven team. You become deeply embedded in a repeatable ecosystem. You aren't bidding on a single localized job. You are betting on a decade -long annuity. of standardized infrastructure work across the country. So what does this all mean for you listening right now? I can hear operators in civil engineering, commercial electrical, and heavy construction thinking, well, I don't forge nuclear control rods or build uranium centrifuges so this market isn't for me. That is a massive misconception about what a modern reactor build actually requires on the ground. It is the single biggest blind spot for regional mega -contractors. People hear the words nuclear reactor, and they
immediately picture a sterile environment. They think of quantum physicists in white lab coats or highly specialized components being welded in a vacuum chamber. Right. But a nuclear reactor build is, first and foremost, a colossal, heavy civil construction project. We're talking about moving mountains of dirt before a single piece of specialized equipment ever arrives. Literally years of heavy civil engineering. You are prepping a site to handle monumental weight loads. There is massive site grading, deep trenching for cooling water intakes, and continuous concrete pours that last for days and require a whole fleet of trucks running in perfect synchronization. It's essentially building a highly secure, self -contained small city. And cities need conventional infrastructure. There are miles of standard mechanical
piping, standard electrical conduit, security fencing, administrative office buildings, access roads, and parking lots for the thousands of workers who will be on site at peak construction. The people moving the dirt, tying the rebar, and pulling the standard copper wire. aren't nuclear scientists. They are regional construction powerhouses. The bulk of the man hours on these sites in the first three years will be executed by traditional heavy civil and commercial contractors. But those traditional contractors still need to pass the pre -qualification gates for the EPC. And this is where the timeline of the $17 .5 billion loan program catches people off guard. The utilities are securing their $500 million equity portions right now. Which brings us to the Monday move. the immediate actionable step
you need to take based on this intelligence. Your directive this week is to look at the landscape and find out which utilities in your region hold those seven letters of intent with Westinghouse. You cannot wait for the official press releases announcing a groundbreaking ceremony. No, absolutely not. By the time the local politicians are holding golden shovels for the cameras, the supply chain is already locked. The heavy civil contracts have been awarded, the laydown yards are paved, and the concrete suppliers are already running test batches. The imperative is to get your firm onto the Westinghouse and the EPC pre -quantification list today. The vetting is happening behind the scenes right now. And frankly, this is where having the right market intelligence is non -negotiable. Using tools like the Smart Business Automator Market Intelligence Pipeline is exactly how forward
-thinking firms are tracking these utility LOIs, monitoring the capital flows, and beginning the NQA1 compliance process before their competitors even realize the game has started. If you wait for a formal request for proposal to hit the public bidding boards, you have already lost. The firms that win this work are the ones who are already pre -qualified and who are sitting at the table helping the EPC define the RFP before it's ever published. And once you're in, you ride that standardized AP1000 wave across the initial five sites and potentially the seven LOI sites after that. For a regional contractor, it is a generational wealth -building opportunity, provided you are willing to endure the pain of that first compliance audit. And the scale of this deployment forces us to look at the long -term macroeconomic impact on the construction
industry itself. This raises an important question for you to consider. If a standardized supply chain and the qualify -once mechanic means a contractor can easily jump from site to site, Does this $17 .5 billion program effectively crown a new, untouchable tier of regional mega contractors? The firms that get on the list today will have a compounding advantage in expertise, compliance, and capital that could allow them to monopolize the next decade of infrastructure. And more importantly, are you going to be one of them, or are you going to let your competitors lock you out? A massive question to chew on as you evaluate your own pipeline and compliance capabilities this week. Get out there, track down those LOIs, and start the pre -qualification process. Thank you for joining us on this deep dive, and we will catch you next time.
