How to Start a Construction Company: The Complete 2026 Blueprint.

A small general contractor sells to four buyer types, and they behave nothing alike. Homeowners doing kitchen, bath, addition and whole-house remodels pay from personal savings, a HELOC or a cash-out refinance, which means their buying decision is rate-sensitive and their payment terms are the best you will ever get: deposit, draws, final on substantial completion. Small commercial tenants and their landlords buy tenant improvement work through a broker or property manager and pay net 45 to 60. General contractors and developers buy you as a subcontractor and pay net 60 to 90 with retainage of 5 to 10% held until closeout. Public agencies pay reliably but require SAM registration, a UEI, and performance and payment bonds costing 1 to 3% of contract value.

The single most important 2026 planning fact is not demand, it is collection speed. Slow payment cost the US construction industry an estimated $280 billion in 2024, and 82% of construction businesses that failed in 2023 cited cash flow problems. That is why the buyer mix you choose determines whether you survive year one. A firm that does $900,000 in homeowner remodels with 30% collected up front is a materially different business from a firm doing $900,000 as a drywall and framing sub for a production builder on net 75 with 10% retainage, even though the top line is identical.

Revenue for a small GC is not a single number. Firms in this segment run from under $500,000 to several million, and the spread is driven by how much you self-perform. A lean firm that subcontracts nearly everything can push $1.5 million to $3 million through two project managers, because your job is coordination, not production. A self-perform outfit with two crews is usually capped near $1.2 million to $2 million until you add a third crew and a dedicated estimator. Industry margins are thin and well documented: roughly 5% average net profit across US construction, with specialty trade contractors around 22.4% gross and best-in-class contractors near 21.8% gross.

The practical read for someone deciding next month: the market will give you work. Residential remodel demand and light commercial TI demand are not the constraint. Your constraints are a license you may not yet qualify for, a bank balance thin enough that one slow-paying customer stalls payroll, and estimating discipline you have not yet built. Do not build the plan around finding customers. Build it around being able to fund the gap between spending on a job and getting paid for it.

A general contractor does not need a storefront and should never pay for one. The real siting question is whether you need a yard. If you subcontract nearly everything, a home office plus a truck works for the first 12 to 24 months, and you save $1,500 to $5,000 per month in rent plus the $3,000 to $10,000 deposit and first month. If you self-perform, you need somewhere to stage material, park two trucks and a trailer, and store $15,000 to $40,000 of tools per crew securely. That means a flex industrial unit, typically 1,500 to 3,000 square feet with a roll-up door, in an industrial-zoned pocket rather than a retail corridor.

Drive time is the hidden cost. If your yard is 35 minutes from your typical job, a four-person crew burns roughly 70 minutes of paid travel per day, which at construction wage levels is real money against a 20% gross margin. Site the yard toward the center of the zip codes you intend to work, not near your house. In dense metros many small GCs skip the yard entirely and use jobsite gang boxes plus a small self-storage unit at $150 to $400 per month, accepting the theft risk in exchange for not carrying a lease.

Your service radius is a licensing question as much as a geography question. Every municipality you pull permits in typically requires its own Business License or Business Tax Receipt at $50 to $300 per year, and some require local contractor registration on top of the state license. Working across six municipalities is cheap in fees and expensive in learning six sets of inspection schedules, plan review timelines and inspector preferences. Most small GCs are better served by dominating three or four jurisdictions where they know the building department staff by name and can get a reinspection scheduled the same week.

One channel decision deserves the same rigor as the physical one: whether you want to be a prime contractor to owners or a subcontractor to larger GCs. Prime work gives you deposits, higher margin and direct customer relationships, but you carry all the coordination risk and the customer's financing risk. Sub work gives you steady volume with no marketing spend, but net 60 to 90 terms and 5 to 10% retainage. Many firms start as a sub to fund payroll and shift toward prime work as their cash position improves.

Take a typical residential remodel contract at $150,000. Direct job cost is labor, materials, subcontractors, permits and equipment rental. At a healthy 22% gross margin, direct cost is about $117,000 and gross profit is about $33,000. The composition matters: on a sub-heavy job, roughly $70,000 to $85,000 of that direct cost goes to subcontractors and the rest to material and your own supervision. On a self-perform job, more of it is your payroll, which means you carry the burden of workers' compensation, payroll taxes and idle time between jobs. Burdened labor cost typically runs 1.25 to 1.45 times the base wage once comp, taxes and holiday time are included.

Now subtract overhead. A small GC's fixed monthly overhead runs roughly $8,000 to $25,000: office or yard rent at $1,500 to $5,000, insurance amortized at $500 to $3,000, vehicle payments and fuel, software at $100 to $500, phones and internet at $300 to $800, plus the owner's or project manager's salary if it is not charged to jobs. If your overhead is $15,000 per month, you need roughly $818,000 in annual revenue at 22% gross margin just to cover it, before any net profit. That single calculation is the one most new contractors never do, and it explains why average construction net profit sits around 5%.

On that $150,000 job, after allocating a proportional share of overhead, a well-run firm keeps $7,500 to $12,000 in net profit, which is 5 to 8%. A firm that underbid by one trade, ate two change orders it never documented, and waited 75 days for the final draw keeps nothing and may be negative. The margin on a single job is not where money is made or lost. It is made or lost in whether the estimate matched the actual, which is why job costing that compares estimated to actual by cost code, on every job, is not optional bookkeeping.

Plan on 18 to 25% gross margin and 3 to 7% net. CFMA benchmarking puts specialty trade contractors near 22.4% gross and best-in-class contractors near 21.8% gross, while average net across US construction sits around 5% and larger public firms land in the 3 to 6% range. If your business plan shows 15% net margin, the plan is wrong, not exceptional.

At opening, the owner is usually the qualifying individual on the license, the estimator, the project manager and the salesperson. That is sustainable for one or two concurrent jobs and stops working at three. The first real hire depends on your model. If you subcontract most work, hire a project manager or superintendent who can run jobs, coordinate subs and handle inspections, at roughly $70,000 to $110,000 per year plus a truck. That hire is what lets you keep selling instead of standing on a jobsite. If you self-perform, hire a working lead carpenter or foreman first, at roughly $55,000 to $85,000, then build the crew of two to four helpers and carpenters beneath that person.

The role almost every new GC skips and then regrets is bookkeeping. A construction bookkeeper who understands job costing, work-in-progress schedules, certificates of insurance tracking, lien waiver collection and certified payroll where applicable costs $30 to $60 per hour part-time or $800 to $2,500 per month outsourced. Generic bookkeeping that produces only a profit and loss statement is nearly useless here, because it tells you the company made money last month without telling you which job lost it. Given that mismanagement is implicated in the large majority of construction failures, this is the cheapest insurance on the list.

Classification is a legal exposure, not an accounting preference. If you direct the schedule, supply the tools and control the method, that worker is an employee, and treating them as a 1099 subcontractor invites a state labor department audit, back payroll taxes and a workers' compensation premium recalculation. Every genuine subcontractor you use must produce a current certificate of insurance and, in licensing states, their own contractor or trade license. Set a rule that no sub steps on site without both on file, and have the bookkeeper track expiration dates.

Safety is a named role even in a small firm. OSHA requires a competent person on site for excavation, scaffolding and fall protection, and in construction that authority must be able to stop work. Designate your superintendent or lead carpenter, pay for the OSHA 30-hour course at roughly $160 to $400 per person, and run a documented weekly toolbox talk. A single recordable injury raises your experience modification factor and can move a $12,000 workers' compensation premium to $18,000 or more for three years.

Your first customers do not come from advertising. They come from the trades you already know. Before the license arrives, call every plumber, electrician, HVAC contractor, roofer and cabinet supplier you have worked with and tell them specifically what you take on and what your service area is. Subcontractors sit in front of homeowners all day and constantly hear about work outside their trade. Follow that with real estate agents who list older housing stock, home inspectors who write reports full of deferred maintenance, and insurance adjusters. Twenty focused conversations in the first three weeks reliably produces more first-year revenue than $5,000 of Google Ads.

The second channel is subcontracting to established general contractors while you build your own pipeline. This is unglamorous and the terms are worse, net 60 to 90 with 5 to 10% retainage, but it fills crew hours immediately and does not require a marketing budget. Call the GCs pulling the most permits in your target municipalities, which you can identify from public permit records at the city or county building department, and ask what they are short of. Framing, punch list and demolition capacity are chronically scarce.

Then build the digital baseline, because it converts referrals rather than generating them. A claimed and photo-filled Google Business Profile, a five-page website with real project photography and license and insurance numbers displayed, and profiles on Houzz and Angi. Budget $1,500 to $5,000 for branding and the site, and treat any paid lead spend as a $500 to $5,000 test with tracked cost per booked job, not a monthly habit. Paid remodel leads commonly run $50 to $200 each with close rates of 5 to 15%, which means a booked job can cost $500 to $2,500 in lead spend. On a 5% net margin, that is only defensible on jobs above roughly $40,000.

Finally, set a rule for job one. Take a project small enough that a 20% estimating error is survivable, roughly $15,000 to $50,000 for a new firm, and require a deposit plus milestone draws in writing. Photograph everything, ask for a written review and three project photos on completion, and collect the referral while the customer is still enthusiastic. Six completed jobs with documented before-and-after photography and six reviews is the asset that makes month four through twelve possible.