Builders usually ask whether they can make more money doing the work themselves, and answer it by comparing the subcontractor's price to their own labor cost — a comparison that puts a fully loaded number on one side and a wage on the other, which is how companies move a trade in-house on the strength of an arithmetic error. This piece reframes the question scope by scope, gives the four tests that decide it, works through a real painting-department comparison including utilization, lays out the hybrid structures that usually beat both extremes, and names the costs that surface six months after the decision.
In this article
- The wrong version of the question
- The four tests
- A worked comparison
- The hybrid arrangements that usually win
- The costs that surface after the decision
- Which trades builders self-perform successfully, and which they should not
- Five mistakes in this decision
- The numbers to track after the decision
- Frequently asked questions
The wrong version of the question
Builders usually frame this as "can we make more money doing it ourselves?" and answer it by comparing the subcontractor's price to their own labor cost. That comparison is almost always wrong, because it puts a fully loaded price on one side and a wage on the other.
The subcontractor's number includes their labor burden, their supervision, their equipment, their insurance, their overhead, their warranty exposure and their profit. Your labor cost, as most builders calculate it, includes the wage and maybe the taxes. Compared that way, self-performing always looks cheaper, and companies move a trade in-house on the strength of an arithmetic error.
The right version of the question is narrower and more useful: for this specific scope, in this company, at this level of volume, which arrangement produces more gross margin per hour of management attention and less risk? That question has different answers for different trades inside the same company, which is why the decision is made scope by scope rather than as a philosophy.
The four tests
- Utilization. Can you keep this crew busy at least 80% of available hours across the year, including the slow season? A self-performed trade is a fixed cost pretending to be a variable one. At 55% utilization, an excellent hourly production rate still produces a losing department, because you are paying for the other 45%.
- Margin. After the full cost comparison, does self-performing produce meaningfully more gross margin per hour than marking up a subcontractor? Marking up a sub is not free money, but it is nearly free of management burden, and a few points of difference does not justify carrying payroll.
- Schedule criticality. Is this the trade that controls your schedule? A scope that everything else waits on — framing on many residential projects, demolition and rough-in on remodels — is worth self-performing at equal margin, because control over the critical path is worth money that never shows up in the trade's own P&L.
- Quality risk. Is this the scope where defects cost you the most, either in rework or in reputation? A builder whose callbacks are concentrated in one trade has a case for bringing it in-house even at a margin penalty, because the rework and the lost referrals are being paid already.
Two yes answers make a strong case. Four make it obvious. One is usually a decision made for emotional reasons — frustration with a subcontractor, most often — and frustration is a bad reason to add a payroll.
Keep reading — free
The rest of this article is worth the 20 seconds
Tell us where to send it and the full piece unlocks right here, along with everything else on this blog.
Rather just talk? Message us on WhatsApp.
A worked comparison
A remodeling company doing $3.2 million a year is deciding whether to bring painting in-house. The subcontractor bills roughly $310,000 a year across their projects, and the company marks that up 18%, producing about $55,800 of margin with essentially no management load.
Self-performing would mean two painters. Fully burdened at $34 per hour each — wage, taxes, workers compensation at the painting class code, benefits and paid time off — that is $141,440 for 2,080 hours each. Add a van at $9,600 a year with insurance and fuel, sprayers, ladders, scaffolding and consumables at $14,000, and supervision at roughly six hours a week of a project manager whose loaded cost is $62 an hour, which is $19,344. Direct cost before materials: $184,384.
Now utilization. The company's own painting demand fills about 3,100 of the 4,160 available hours, which is 75%. The remaining 1,060 hours are either idle or sold to other builders — and selling them requires a sales effort the company does not currently have. At 75% utilization the effective cost per productive hour is $59.48, and the work that would have gone to the sub at $310,000 now costs $184,384 plus materials.
The arithmetic looks favorable until the idle hours and the management time are weighted honestly, and until someone asks what happens in the quarter when three projects slip and painting has nothing to paint. The company's actual decision was to self-perform a narrower scope — punch-list painting, small repairs and touch-ups, the work that was hardest to get a sub to show up for — with one painter, and keep the sub for full-project work. That hybrid answer is the most common right answer and the one the binary question hides.
The hybrid arrangements that usually win
Very few companies should self-perform everything or nothing, and several intermediate structures capture most of the benefit with less exposure.
Self-perform the small and the awkward. Punch work, warranty calls, small repairs and the scopes too small for a subcontractor to schedule promptly. These are the jobs that damage client relationships when they wait three weeks, and having one or two capable people in-house solves a chronic problem at modest cost.
Self-perform the critical path and sub the rest. Bring in-house only the trade that everything waits on, keeping the rest variable. This buys schedule control without turning the company into an employer of five trades.
Keep a core crew and flex with subs. A small permanent crew sized to the slow season, supplemented by subcontractors at the peak. It sacrifices some peak margin in exchange for never carrying idle payroll, which is the failure mode that closes construction companies.
Labor-only subcontracting. The company buys materials and manages the work while a subcontractor supplies crews. It captures material margin and schedule control without payroll, though the classification and insurance implications need to be handled properly rather than casually.
The costs that surface after the decision
Three consequences of self-performing show up six months later and are rarely in the original comparison.
Management attention is finite, and a self-performed trade consumes it disproportionately. Hiring, training, scheduling, performance problems, tool accountability and the daily supervision of a crew all draw on the same owner or project manager who was previously coordinating a subcontractor with a phone call. Companies that added a trade often report that the trade was profitable and everything else got slightly worse.
Insurance and classification change. Adding a trade means new workers compensation class codes at that trade's rate, potential changes to general liability, and in some cases a licensing requirement in that specialty. These are knowable in advance and are frequently discovered at the audit instead.
And the exit is harder than the entry. Subcontractor relationships that lapse take a year to rebuild, and the good subs fill their calendars with builders who stayed loyal. A company that self-performs a trade for two years and then wants to go back frequently finds the pricing worse than it was.
Which trades builders self-perform successfully, and which they should not
Patterns repeat across the industry, and they follow the four tests rather than tradition. Knowing where your candidate scope sits saves a year of learning it firsthand.
The trades most commonly self-performed successfully share three characteristics: they appear on nearly every project, they require capability rather than licensure, and their equipment is modest. Carpentry and framing, demolition, general labor and cleanup, drywall and painting in some markets, and concrete flatwork for companies with steady volume all fit that description. They also tend to sit on the critical path, which is what makes the schedule argument work.
The trades that punish self-performing share the opposite characteristics: specialty licensure, expensive equipment, high liability exposure, and demand that is intermittent on any single builder's project list. Electrical, plumbing, mechanical, roofing, elevator work and fire protection generally belong here. The licensure alone changes the calculation, because the company now depends on retaining a specific licensed individual, and losing that person can halt work rather than merely inconvenience the schedule.
There is a middle group where the answer genuinely depends on the company: excavation and sitework, where equipment cost is the deciding variable; tile and finish carpentry, where the quality argument is strongest and the volume argument is weakest; and low-voltage and cabinetry installation, which sit at the boundary of what a general crew can absorb.
One caution about the middle group. A builder who self-performs a trade at low volume tends to keep the least experienced people in it, because the good ones get pulled to whatever is urgent. That is the mechanism by which a department created to improve quality ends up producing the callbacks it was meant to prevent — and it is a staffing consequence, not a strategy failure, which means it is fixable if it is anticipated.
Five mistakes in this decision
- Comparing a sub's price to a wage. One number is fully loaded and the other is not, so self-performing always wins on paper.
- Ignoring utilization. A crew busy 55% of the year turns good production rates into a losing department.
- Deciding out of frustration. A difficult subcontractor is a sourcing problem, not a reason to add payroll.
- Treating it as all or nothing. The hybrid answers — small scopes, critical path only, core crew plus flex — usually beat both extremes.
- Forgetting the exit cost. Subcontractor relationships that lapse are expensive to rebuild, and the good ones will be busy.
The numbers to track after the decision
Five figures tell you whether a self-performed trade is working: crew utilization by month, which is the number that determines everything else and the one that deteriorates first; gross margin per productive hour, compared against the markup you were earning on the subcontractor; rework and callback rate for that scope before and after bringing it in-house, since quality was often the stated reason; management hours per week consumed by the trade, which quantifies what it is taking from the rest of the company; and schedule performance on the projects where that trade is on the critical path, which is the benefit most often claimed and least often measured.
Run the comparison properly for one of your trades
Send us one trade's annual subcontracted volume, your markup on it, and your estimate of the hours it would require. We will build the fully loaded comparison including utilization and management time, and tell you whether self-performing, a hybrid or staying with the sub produces more margin at your volume.
Frequently asked questions
How should you compare self-performing against subcontracting?
Apples to apples, which almost never happens. The subcontractor's price includes their labor burden, supervision, equipment, insurance, overhead, warranty exposure and profit; the builder's labor cost, as usually calculated, includes the wage and maybe the taxes — compared that way self-performing always looks cheaper. Against the sub's price you have to put wages, payroll taxes, workers compensation at that trade's class code, benefits and paid time off, tools and consumables, equipment and maintenance, vehicle and fuel, supervision time, training, warranty and rework, and the idle hours between jobs when the crew is on payroll with nothing to do. That last line decides most of these decisions and is the one almost nobody includes. The useful question is narrower: for this scope, at this volume, which arrangement produces more gross margin per hour of management attention and less risk?
What are the tests for deciding whether to bring a trade in-house?
Four. Utilization: can you keep the crew busy at least 80% of available hours across the year including the slow season, since a self-performed trade is a fixed cost pretending to be variable and at 55% utilization an excellent production rate still produces a losing department. Margin: after the full comparison, does self-performing produce meaningfully more gross margin per hour than marking up a sub, given that the markup is nearly free of management burden. Schedule criticality: is this the trade everything else waits on, because control over the critical path is worth money that never appears in that trade's own P&L. And quality risk: is this the scope where defects cost the most in rework and reputation. Two yes answers make a strong case, four make it obvious, and one usually means the decision is being driven by frustration with a subcontractor.
What does a realistic self-perform comparison look like in numbers?
Take a remodeling company at $3.2 million a year deciding on painting. The sub bills about $310,000 annually, marked up 18%, producing roughly $55,800 of margin with essentially no management load. Self-performing means two painters fully burdened at $34 per hour — wage, taxes, workers compensation at the painting class code, benefits and paid time off — which is $141,440 for 2,080 hours each, plus a van at $9,600 with insurance and fuel, sprayers, ladders and consumables at $14,000, and supervision at six hours a week of a project manager at $62 loaded, which is $19,344: $184,384 before materials. Then utilization: the company's own demand fills about 3,100 of 4,160 available hours, or 75%, making the effective cost $59.48 per productive hour. The company's actual decision was a hybrid — one painter for punch and small repairs, sub for full projects.
What hybrid arrangements work between self-performing and subcontracting?
Four, and very few companies should be at either extreme. Self-perform the small and awkward — punch work, warranty calls, small repairs and scopes too small for a sub to schedule promptly, which are exactly the jobs that damage client relationships when they wait three weeks. Self-perform the critical path and sub the rest, bringing in-house only the trade everything waits on, which buys schedule control without becoming an employer of five trades. Keep a core crew sized to the slow season and flex with subs at the peak, sacrificing some peak margin to avoid ever carrying idle payroll. And labor-only subcontracting, where the company buys materials and manages the work while a sub supplies crews, capturing material margin and schedule control without payroll — though classification and insurance implications need proper handling rather than casual treatment.
What costs appear after deciding to self-perform a trade?
Three, and they show up around month six. Management attention is finite, and a self-performed trade consumes it disproportionately through hiring, training, scheduling, performance problems, tool accountability and daily supervision — drawing on the same person who previously coordinated a subcontractor with a phone call, which is why companies often report that the new trade was profitable and everything else got slightly worse. Insurance and classification change, with new workers compensation class codes at that trade's rate, possible general liability changes and sometimes a specialty licensing requirement, all knowable in advance and frequently discovered at the audit. And the exit is harder than the entry: subcontractor relationships that lapse take a year to rebuild, and the good subs fill their calendars with builders who stayed loyal.