How do you start a custom welding fabrication business in 2027?

What custom fabrication actually is, and why the 2027 window is open

Custom welding fabrication is the business of turning raw steel, stainless, and aluminum into one-off or low-volume assemblies that do not exist as a catalog part: a stair stringer cut to fit a shaft that was poured out of square, a sanitary hopper for a brewery scaling from a 15bbl to a 30bbl system, a cracked excavator bucket rebuilt overnight, a sculptural railing drawn by an interior designer on a napkin. It is distinct from production welding — a shop running the same weldment 4,000 times — and that distinction is the whole strategic point in 2027, because production welding is exactly what automation is absorbing and custom work is exactly what it cannot touch.

Three structural forces make this a genuinely strong business to start right now, and it is unusual for a small business to sit on all three at once.

The labor shortage is real and quantified. The American Welding Society has for years published projections of a welder shortfall in the hundreds of thousands, with figures in the range of 320,000–360,000 unfilled positions by 2027 as the existing workforce ages out — the average welder is in their mid-to-late 50s — and trade-school enrollment has never caught up. A shortage in your trade is a pricing-power gift. When a general contractor cannot find a misc-metals sub who answers the phone, the conversation stops being about price and starts being about availability. That is a different conversation, and it is the one you want to be in.

Fabrication is a real-asset, hard-to-offshore business. You cannot email a 400-pound stainless tank or a two-flight steel stair to an overseas plant and have it arrive tomorrow, square, and to the field dimensions you took last Thursday. Custom metal work is inherently local, inherently logistics-bound, inherently relationship-driven. The freight cost of steel and the physical necessity of field-measuring create a moat that no amount of cheap offshore labor erodes.

Reshoring and infrastructure spend are tailwinds through the late 2020s. Domestic manufacturing buildout, data-center construction, grid hardening, energy projects, and the long tail of infrastructure funding all generate structural and miscellaneous-metals demand, and a meaningful share of it flows downhill to small shops as subcontract packages. You do not have to win the megaproject; you have to be the shop the megaproject's GC calls for the embeds and the stair rails.

The counter-trend is worth naming honestly. Cobot and robotic welding cells keep getting cheaper and easier to deploy, and they are eating high-volume repetitive production welds. Simultaneously, a flood of hobbyist-grade equipment has lowered the barrier to entry at the bottom of the market — anybody can buy a flux-core machine and put "welding" on a truck. Neither of those touches the defensible middle. A cobot cannot walk a job site and fab a stringer to fit reality. A hobbyist cannot pull a city business license, carry $2M in liability with the GC named as additional insured, produce weld certs and qualified procedures, and hit a construction schedule. The winning 2027 posture is explicitly custom, certified, and niched — the part of the market that is growing and defensible, not the part that is shrinking and commoditized.

The market structure supports a new entrant. US metal fabrication is a large, deeply fragmented industry — tens of billions in annual revenue spread across roughly 60,000-plus establishments, the great majority of which employ fewer than 20 people. There is no dominant national player in custom fab the way there is in package delivery. That fragmentation is your opening: a well-run, well-niched shop can take share in a metro without ever fighting a giant. But treat the national number as a vanity metric. What matters is your serviceable slice — the custom and misc-fab spend within a 60–90 minute drive, in your chosen niche, which in a typical mid-size metro commonly runs $15M–$60M a year. What one new shop realistically captures in Years 1–3 is $150K–$700K, a fraction of one percent of that. That math should encourage you. You do not need to win the market. You need twenty to fifty good customers.

Segment the demand by who signs the check, because the four buyer types behave completely differently. General contractors and construction managers buy misc and structural metals as subcontract packages — the largest, steadiest pool, but net-30 to net-60 and bid-competitive. Manufacturers and processors — breweries, food plants, equipment OEMs — buy custom weldments as production support and capital projects: higher margin, stickier, but gated on capability and certs. Ag, trucking, and construction-equipment operators buy repair-plus-fabrication: smaller tickets, fastest cash, fiercely loyal. Architects, designers, and custom-home builders buy ornamental and architectural metal: best margins and brand story, most design-iteration overhead. Homeowners buy one-off repairs — fine as Year 1 fill-in work, terrible as a primary segment because it is low-ticket and high-tire-kicker.

One more framing worth internalizing before any of the mechanics: the thing that separates a fabrication *business* from a fabrication *job* is not skill at the hood. It is the same set of levers any operator would recognize — a repeatable offer, a pricing model, a pipeline, a quote-to-cash cycle, and someone other than the founder producing the work. Fab founders who think of themselves as metal people rather than operators consistently underinvest in exactly those levers, and it caps them at $200K forever.

The step-by-step process from decision to first invoice

The sequence matters. Founders who buy iron before they pick a niche end up with the wrong iron, and founders who sign a lease before validating demand pay rent on an empty shop.

Step 1 — Pick the primary wedge. Do this before you spend a dollar. There are four strong 2027 wedges.

*Structural and miscellaneous metals for GCs*: stairs, railings, handrails, ladders, bollards, embeds, lintels, gates, mezzanines. Huge steady demand, clear repeat customers because the same GCs bid job after job, and stairs and rails systematize into near-products. Downside: net-30/net-60 terms, bid competition, drawing literacy required, constant field measuring. Solo-to-small-crew range: $180K–$650K.

*Food-grade and sanitary stainless TIG*: breweries, distilleries, wineries, dairies, food processors, extraction operations — tanks, hoppers, conveyors, platforms, sanitary piping, custom equipment. Highest margins in the trade, stickiest relationships, and certs plus capability create a real moat because contamination is a business risk to the customer. Downside: TIG mastery and sanitary discipline are non-negotiable, equipment investment is higher, first customers come slowly. Range: $250K–$900K.

*Trailer, equipment, and heavy-truck repair-plus-fab*: cracked frames, broken buckets, custom hitches, equipment modifications, trailer builds. Steadiest cash flow in the trade because repair gets paid fast, intensely loyal customers, and downturn-resistant since people repair more when money is tight. Downside: dirty, physical, mobile-rig demand, smaller average tickets. Range: $200K–$500K.

*Architectural and ornamental*: feature stairs, sculptural railings, gates, fireplace surrounds, furniture, signage. Best brand and margin story, portfolio-driven marketing, design-build premium. Downside: design-iteration overhead, long sales cycles, lumpy project revenue, brutal finish-quality bar. Range: $150K–$400K.

Step 2 — Validate with actual conversations. Call fifteen to twenty-five potential buyers in your wedge before you commit. Ask GC estimators who currently does their misc metals and what frustrates them. Ask head brewers who fabbed their last platform. Ask a fleet manager how long they wait on repairs. You are testing two things: is there unmet demand within driving distance, and what specifically is the incumbent doing badly. That second answer becomes your positioning.

Step 3 — Form the entity and stack the paper. LLC at minimum — never operate as a sole proprietor in a trade where you weld structural elements and move heavy steel. Get the EIN, the business bank account, the city and state business licenses, and check whether your state requires contractor licensing for the work in your wedge (structural and permitted work often does, and stamped drawings mean coordinating with a licensed engineer).

Step 4 — Bind insurance before the first job. General liability at $1M–$2M is the floor; many GCs require $2M and want to be named as additional insured before you set foot on site. Add commercial property/equipment, commercial auto, and workers' comp the moment you have an employee — in many states, effectively for yourself as well.

Step 5 — Secure certs that unlock revenue. AWS certifications and customer- or code-specific qualifications are not legally required to operate, but they gate the good work. D1.1 structural quals for Wedge 1; TIG and sanitary quals for Wedge 2. Treat these as revenue investments, not compliance chores.

Step 6 — Lease the space. Light-industrial zoning with confirmed permitted use, 200–400 amp three-phase service, a floor that takes the load, ceiling height for a crane, and doors big enough to move finished assemblies. Confirm all of it before signing.

Step 7 — Buy the iron. Used and cash on the core kit, financed on the CNC plasma table. Details in the cost section below.

Step 8 — Build the quote template and the shop rate model before the first inquiry lands, so you are never quoting from scratch under pressure.

Step 9 — Land the first three jobs and treat them as auditions. For a GC, the first three quotes and schedules determine whether you are on the bid list for the next five years. For a brewery, the first job builds the portfolio that the entire brewer network will hear about. Photograph everything.

Step 10 — Invoice the day the job is done and collect. Then follow up a week later, which is how a job becomes a relationship.

The operational cycle inside that last box, once you are running, is: inquiry arrives → qualify fast against your niche and size filters → field measure or review drawings → quote with clear scope, exclusions, lead time, and payment terms → win and confirm with a signed quote or PO plus a 30–50% deposit on non-GC project work → order material the day the job is confirmed because steel lead times and pricing move → schedule with realistic buffers → fabricate with QC checkpoints at fit-up and after welding → finish or send to your powder coater or galvanizer → deliver or install → invoice immediately → collect and chase the moment it ages past terms.

Shops bleed money in exactly two places in that cycle: quoting that is too slow, too inconsistent, and full of scope gaps that become unpaid change orders; and collections that have no deposit, no same-day invoicing, and no receivables discipline. Fix those two and the rest of the workflow largely runs itself.

Costs, timelines, and the numbers that decide whether it works

The lean build: $35K–$55K. Used multiprocess welder $2.5K–$5K. Used TIG machine $1.5K–$3.5K. Hand plasma cutter $800–$2K. Used ironworker or small press brake/shear combo $4K–$10K. Used drill press and band saw $1.5K–$3K. Grinders and hand tools $2K–$4K. Fixturing and welding tables $1.5K–$4K. Used gantry or engine hoist for lifting $1K–$3K. PPE and ventilation $1.5K–$3K. Used work truck or trailer $5K–$15K. First, last, and deposit on a 1,500–2,500 sq ft lease $4K–$10K. Cylinder leases and first fill $600–$1.5K. Insurance down payment $1.5K–$4K. Entity and legal $800–$2K. Working capital for early steel buys and the net-30 gap $5K–$12K.

The serious build: $70K–$120K. New or low-hour multiprocess and TIG machines, an entry CNC plasma table at $18K–$40K, a real press brake and shear, a two-ton overhead crane on a freestanding gantry or bridge, 3,000–4,000 sq ft, a better truck, and three to four months of operating reserve. Rent for a fab-rated commercial space typically runs $1.2K–$4.5K/month for 1,500–4,000 sq ft depending on metro.

Buy, finance, or lease — the framework matters as much as the totals. Buy used and pay cash for welding machines (a well-maintained major-brand multiprocess off the used market is 95% of a new one at 50–60% of the price, and welders hold value), basic shop tools, fixturing, hand tools, and lifting gear. Finance the CNC plasma table and larger forming equipment — the table is the single highest-leverage purchase a fab shop makes, because it converts you from cut-by-hand to parts-on-demand, sharpens quote speed and accuracy, and lets a helper produce parts. Amortizing it over three to five years against the revenue it generates is sound; paying cash and starving working capital is not. Lease and never buy gas cylinders — ownership is a hassle and a capital trap. And do not buy a building in Years 1–3; premature real estate is the most common way fab founders trap their capital and lose flexibility.

Pricing is where most shops kill themselves. The instinct is to price off your old wage: "I made $32/hour as an employee, so $55/hour is great money." It is wrong and it is a death spiral. Your fully-loaded shop rate has to cover rent, utilities, equipment depreciation and replacement, consumables and gas, insurance, truck, software, marketing, accounting, bad-debt risk, profit, and — critically — your non-billable hours quoting, ordering, invoicing, and sweeping the floor. Build the model honestly and a solo shop needs to *bill* $85–$165/hour just to net what a good welding job pays as an employee, before you pay anyone else.

Three approaches. Hourly shop rate is transparent but caps your upside — get faster and the customer captures the savings — and invites customers to police your clock. Per-piece quoting is better and is how you should price by Year 2: quote the deliverable, and your speed becomes margin. Cost-plus with material markup and a labor rate is what GCs expect on bid work. Mature shops use per-piece for known repeat work, cost-plus for bids, and pure hourly only for diagnostic repair where scope is genuinely unknown.

Mark up material 15–35%. You carry the cost, the procurement labor, the storage, the cut-off waste, and the risk. Customers who fight material markup are telling you in advance that they will be a problem.

A worked example. A $4,000 stair-and-rail job: material costs you $1,100 and bills at $1,400, consumables and gas $120, labor 22 hours. Inside the $4,000 fixed price your effective rate lands near $116/hour after material margin — healthy. Run the same job underpriced at $2,600 and the arithmetic collapses: material still costs $1,100, you net roughly $1,500 across 22 hours, about $68/hour, and you cannot hire anyone at that rate. Target gross margin for a healthy custom fab shop is 35–55% (food-grade highest, bid structural lowest) with net margin for a well-run solo-to-small shop at 18–32%. Budget 3–6% of revenue for consumables and gas, and track it — this is the line item founders consistently forget.

The trajectory. Year 1: solo or solo-plus-helper, $120K–$280K revenue, 50–60 hour weeks, too much non-niche work, learning to quote, owner take $50K–$110K. The goal of Year 1 is to survive, learn, and build the relationship base. Year 2: one to two employees, $220K–$420K, niche now 50–70% of revenue, quoting fast and profitable, one price increase behind you, owner take $80K–$160K. Year 3 is the inflection: two to four employees, $350K–$700K, founder welds less than half the time, systems exist, owner take $120K–$250K. This is where it becomes a real business or stalls as an over-busy job. Years 4–5 for a lifestyle shop: five to twelve employees, $900K–$2.2M, founder mostly out of production, net margin 15–25%, owner take $200K–$500K plus a sellable asset. The alternative is productizing — developing your own line (a trailer model, mezzanine kits, a furniture line) sold beyond the local market, which is harder and more capital-intensive but lifts the ceiling to $3M–$8M and builds a far less owner-dependent company.

Software and back office. Budget $150–$600/month for the whole stack and stop treating it as optional. Quoting and estimating is the highest-leverage software in the building — spreadsheet templates are fine in Year 1, dedicated fab-estimating or shop-management tools after that; the goal is turning a quote around in hours rather than days. 2D CAD for shop drawings, plus CAM and nesting once you have a plasma table (often bundled). Accounting on QuickBooks Online, with a bookkeeper by the time you hit $250K. Job tracking so nothing falls through and GC relationships survive. Invoicing that takes cards and ACH. A lightweight CRM to track GC and processor relationships, bid history, and follow-ups — the same discipline any RevOps operator would apply to a pipeline, just applied to twenty-five relationships instead of two thousand.

Financing the startup. Most founders blend sources. Personal savings is the cleanest base but do not drain your entire reserve buying iron. Equipment financing is the right tool for the plasma table and forming gear — the equipment is the collateral and you match the cost of the asset to the revenue it generates. SBA loans (7(a) for general startup and working capital, 504 for real estate and major equipment) are accessible to fabrication businesses and can fund a more serious build, with a solid plan, owner equity, and patience for the process. A line of credit is the most underrated tool here — not to fund the startup, but to bridge steel-and-labor outflow against net-60 collections. Establish it while the business looks healthy, because banks extend credit when you do not need it and refuse when you do. Avoid funding an asset-heavy, thin-early-margin business on high-interest cards beyond short bridge use.

Where founders get it wrong

The default-playbook trap. The most common failure is not bankruptcy, it is never becoming a real business. The default playbook: a skilled welder gets tired of working for someone else, buys a machine and a truck, hangs a shingle that says "[Name] Welding & Fabrication," and takes every job that walks in — a handrail this week, a trailer repair next, an art sculpture after that. It feels like freedom because the phone rings and the work is varied. Here is the mechanism that makes it a trap: when you take any job, you cannot build a system. Every job is new material, new process, new fixture, new learning curve, new quote from scratch, new customer with no repeat potential. You never climb a learning curve because you never do the same thing twice. Your quoting stays slow and inconsistent because you have no reference jobs. You cannot train a helper because nothing is repeatable — it all lives in your head. You cannot market because your message is "I weld things," which is indistinguishable from every competitor. And you cannot raise prices because with no positioning you are a commodity, and commodities compete on price. Every shop that broke $500K niched. The hardest part is the fear of saying no, and in Year 1 you will say yes more than you should — that is fine. But you must have a stated niche and the percentage of revenue coming from it must climb every quarter.

Cash-flow whiplash. This kills otherwise-healthy shops and deserves its own attention. You pay for steel and gas at or near cash terms, you pay welders weekly, rent and utilities are monthly — and your customers, especially GCs, pay net-30 at best and net-60 to net-90 in practice. You are financing your customers. A growing shop is chronically cash-poor *because* it is growing: every new job ties up cash in material and labor before the invoice gets paid. The defenses: collect 30–50% deposits on all non-GC project work and progress-bill long jobs; invoice the day the job is done, not at month-end; earn net-30 supplier credit as soon as you have a track record, since supplier terms directly offset customer terms and are the cheapest working capital you will ever get; chase receivables relentlessly, because the squeaky shop gets paid first; understand mechanic's lien rights and file preliminary notices where they apply; keep two to four months of fixed costs in reserve; and fire slow-pays — a customer at net-90 is not a good customer regardless of volume. Many founders find the most profitable thing they do in Year 2 is not landing an account but fixing collections.

Owner-as-only-welder dependency. As long as you are the only person who can do the work, the business is capped at your two hands and it stops when you take a vacation or get hurt. The first hire is a shop helper or apprentice who can cut, grind, prep, fixture, and run the plasma table — the roughly 60% of shop work that is not the critical weld. Make it the moment you are consistently turning away work or working past 55 hours a week. The second hire is a second skilled fabricator so two jobs run in parallel. The third, often part-time first, is office and admin, because the founder doing $25/hour admin work is the most expensive labor in the building. Hiring is hard in a shortage: build a pipeline with the local community college welding program, hire apprentices and train them up, pay above market, keep a clean well-equipped shop (good welders choose shops with good gear), and offer a path — a raise ladder, profit share, eventually a buy-in. Treat welders as the scarce resource they are or your competitor will.

Customer concentration. If one GC is 40% of revenue and they have a bad year or switch subs, you are in trouble. Keep no customer above 25–30%.

The rest of the Year 1 list. No deposits. Slow, inconsistent quoting. Buying everything new. Buying a building too early. No QC discipline — one bad weld on a structural job ends a GC relationship. Scope gaps in quotes, which become unpaid change orders. Skipping certs and insurance, which are revenue enablers rather than costs. Underestimating consumables. No operating reserve, so one slow month becomes a crisis. Eating steel price swings instead of putting 15–30 day validity windows on quotes. Running a single-machine shop on a critical process with no preventive maintenance and no relationship with a repair tech.

Quality and reliability are the actual marketing engine. Quality is non-negotiable because in a tight local network a visible flaw is not a complaint, it is a story that travels. Build QC in from day one: fit-up check before welding, visual inspection of every weld, dimensional verification against the drawing, and the documented procedures that code and food-grade work require. Write checklists for repeat job types and use them — the checklist is what lets a helper produce work to your standard. Reliability is the other half and often the half that wins the long-term customer: a GC forgives a higher price for a sub who answers the phone, quotes fast and clean, shows up when they said, and never blows a schedule. Reliability is a system — job tracking, buffered scheduling, proactive communication when something slips, and the discipline to not take more than the shop can deliver. Shops with that reputation stop bidding against five competitors and start getting the call directly. Becoming the default beats being the cheapest.

Decision framework: choosing your wedge, your pricing, and your next hire

Run yourself through the gates honestly before signing anything.

Skill gate. Are you genuinely good at the welding and fabrication your chosen niche demands — food-grade TIG, structural, repair, ornamental — or close enough to get there in months? If not, get the skill first, on someone else's payroll.

Capital gate. Can you assemble $35K–$120K through savings, an SBA or equipment loan, or a partner, *and* survive six to twelve months of lean income? If not, build the runway before you build the shop.

Niche gate. Have you picked a primary wedge, and is there validated demand within driving distance? Validate by calling GCs, brewers, and fleet managers — not by assuming.

Business-temperament gate. Are you willing to quote, sell, collect, hire, and manage, or at least learn those and eventually delegate them? Shops that stall are run by people who only want to weld.

Lifestyle gate. Are you prepared for the physical demand and the Year 1 grind, with a support system for it? The work is hot, loud, dirty, and hard on a body over decades.

Exit-vision gate. Do you know what you are building toward — a lifestyle shop, a productized company, a sellable asset — so today's decisions ladder toward it? Owner-dependency is what sets the multiple: a shop where the founder is the only skilled hand and every relationship runs through them trades at roughly 2–3x SDE because the buyer is purchasing a job with equipment attached. A shop with a trained crew, documented systems, diversified customers, and a founder out of production trades meaningfully higher, commonly 3–5x SDE, and a productized fabrication company with a real product line and a management team can attract strategic or private-equity interest at 4–6x-plus EBITDA. The single most valuable thing you can do for an eventual exit is the same thing that makes the business good to own: get out of the hood, build a crew, document the systems, diversify the customers.

The location and layout decision. Light-industrial on the metro edge gives cheaper rent, fewer noise and zoning headaches, and truck access, at the cost of visibility and longer drives to job sites — which matters most for Wedges 1 and 3, where you live on site. For most niches Year 1, take light-industrial with good highway access; your customers find you by relationship, not by driving past. Confirm permitted use and power before signing: a serious shop wants 200–400 amp three-phase, and retrofitting power is expensive. Lay the shop out so flow runs one direction — receiving and steel storage near the overhead door, then cutting, then forming, then fit-up and welding, then grinding and finishing in a separate area so dust does not contaminate stainless prep, then staging near the out door. Cross-traffic costs hours a day. Lease slightly bigger than you need; a 1,500 sq ft shop feels tight fast, and re-leasing in eighteen months is worse than paying a little more now.

The sourcing decision. Build relationships with two or three steel service centers, not one — you want competitive pricing, backup when a size is out, and the leverage of being a known account in multiple places. Order full lengths and standard sizes where practical, since cut-to-size costs more per pound; keep a working inventory of your most-used sizes; and track your drop and offcut inventory, because a fab shop accumulates usable material that fills small jobs at near-zero material cost. Check lead times before quoting — stainless and specialty alloys are rarely same-day, and a job quoted without confirming availability is a schedule blown waiting on steel. Build the same partner relationships with your gas supplier, powder coater, galvanizer, machine shop, and equipment repair tech; they are operational necessities and referral sources both. The powder coater you feed sends fab work back.

The revenue-mix decision. Fabrication revenue is uneven in two ways. Seasonally, construction work slows in deep winter in cold climates, ag work follows the farming calendar, and architectural work is lumpy year-round. Cyclically, construction and capital projects track interest rates and the broader economy while repair work is comparatively countercyclical, since customers repair instead of replace when money tightens. A shop that is 100% commercial-construction misc metals is a leveraged bet on one cycle. Pairing a construction-driven primary with a repair-work secondary builds in a shock absorber. Watch the leading indicators — GC payment behavior stretching out, bid volume dropping, quote-to-win rates falling — and tighten cash and pipeline before a slowdown lands rather than after.

Where the work comes from. In rough order of value: direct relationships with GC project managers and estimators; presence in your niche's industry community (brewers know brewers, and one great brewery job referred well becomes five); word-of-mouth in the trade for repair work; a real website with a real portfolio, not a social page, because GCs and designers will look you up; Google Business Profile and local search, since "[city] metal fabrication" is genuine buying intent; strategic supplier relationships, because your steel and gas suppliers get asked "who does good fab work?"; and referrals from adjacent trades. Paid advertising works poorly for custom fab, with targeted local search the exception. The brand discipline that matters most is being relentlessly consistent about your niche — a shop whose website, truck, and every conversation say "we do food-grade stainless, nothing else" gets filed under exactly that. A shop that says "we weld anything" gets remembered for nothing.

Related questions

How much does it cost to start a small welding shop?

A lean build runs $35K–$55K covering used machines, basic forming equipment, tools, PPE, a truck, lease deposits, insurance, and working capital. A serious build with a CNC plasma table, press brake, overhead crane, and 3–4 months of reserve runs $70K–$120K.

Do you need certification to start a welding business?

No certification is legally required to form the business, but AWS D1.1 structural quals and sanitary/TIG qualifications gate the profitable work. GCs, food processors, and any code-governed project expect certified welders and qualified procedures before awarding.

What hourly rate should a fabrication shop charge?

$85–$165/hour fully loaded, depending on niche and market — not $45, which is a race to the bottom. Better still, quote per-piece by Year 2 so your speed becomes margin instead of a discount handed to the customer.

How long until a welding fabrication business is profitable?

Most shops are cash-positive within months but the owner take is modest in Year 1 at $50K–$110K on $120K–$280K revenue. Real profitability arrives in Year 3 at $350K–$700K revenue with the founder welding less than half the time.

Is welding a good business to start given automation?

Yes, for custom work. Cobots absorb high-volume repetitive production welds, which pushes human value toward one-off, field-fit, complex-geometry, judgment-heavy fabrication that automation cannot do. Niched custom shops sit on the right side of that trend.

FAQ

Should I start mobile or lease a shop?

A mobile rig is the cheapest entry and works well for repair-oriented Wedge 3, letting you generate revenue before committing to rent. But mobile alone caps you: you cannot fixture large assemblies, run a plasma table, or store steel. Most founders who start mobile add a shop within twelve to eighteen months. If your wedge is structural, food-grade, or architectural, lease from the start — those customers expect a shop.

What is the single highest-leverage equipment purchase?

An entry CNC plasma table at $18K–$40K. It converts you from a cut-by-hand shop to a parts-on-demand shop, dramatically improves quote speed and accuracy because you can predict cut time, and lets a helper produce parts without your supervision. Finance it over three to five years against the revenue it generates rather than paying cash and starving working capital.

How do I get on a general contractor's bid list?

Visit the project managers and estimators in person, ask what their current misc-metals sub does badly, and offer to quote something small. Submit fast, professional, itemized quotes with clear exclusions. Carry the insurance limits they require and be willing to be named as additional insured. The first three jobs are auditions — hit the schedule and the quality bar and you are on the list for years.

How much working capital do I need beyond equipment?

Enough to cover two to four months of fixed costs plus the material float on jobs in progress. For a lean solo shop that is typically $5K–$12K at minimum; for a shop with employees and GC receivables at net-60, plan on materially more and establish a line of credit while the business looks healthy rather than when you need it.

Can I run this part-time while keeping a job?

Some founders do for the first six to nine months, taking evening and weekend repair work to build a customer base and buy iron without debt. It works for Wedge 3 repair and small architectural jobs. It does not work for GC structural work, where schedule responsiveness during business hours is the product you are selling.

What does the shop sell for when I want out?

An owner-operator shop where the founder is the only skilled hand typically trades at 2–3x SDE, because the buyer is purchasing a job with equipment. A shop with a trained crew, documented systems, and diversified customers trades higher, commonly 3–5x SDE. A productized company with a real product line and a management team can reach 4–6x-plus EBITDA and attract strategic or private-equity buyers.

Sources

• https://www.aws.org/ — American Welding Society: certifications, qualification standards, and workforce research
• https://www.bls.gov/ooh/production/welders-cutters-solderers-and-welding-fitters.htm — Bureau of Labor Statistics occupational outlook for welders and fabricators
• https://www.sba.gov/funding-programs/loans — SBA loan programs including 7(a) and 504
• https://www.osha.gov/welding-cutting-brazing — OSHA welding, cutting, and brazing safety standards
• https://www.census.gov/naics/ — NAICS classification and industry definitions (fabricated metal product manufacturing, 332)
• https://www.thefabricator.com/ — The FABRICATOR: metal fabrication industry trade publication
• https://www.irs.gov/businesses/small-businesses-self-employed — IRS small business and self-employed tax guidance
• https://www.score.org/ — SCORE: free small business mentoring and planning resources
• https://www.aisc.org/ — American Institute of Steel Construction: structural steel standards and certification

Related on PULSE

• How do you start a mobile welding business?
• How do you price custom fabrication jobs profitably?
• How do you land your first general contractor as a subcontractor?
• How do you hire skilled trades workers during a labor shortage?
• How do you manage cash flow with net-60 customers?
• How do you productize a service business into a product line?