How to find new customers for a machine shop or fab shop
Referrals, trade shows, reps, directories, cold outreach, SEO and Google Ads compared for job shops: cost, speed and the kind of customer each one brings in.
Frank Frenna & Gianluca Amato11 min read
Ask most machine shop and fab shop owners where their customers come from and the answer is the same: referrals, repeat work, and one or two big accounts. That's a healthy base. The risk is concentration. When your largest customer moves a product line, changes buyers or brings work in-house, a third of your revenue can disappear in a quarter, and there's no system to replace it.
This guide compares the ways job shops actually find new customers, what each one costs, how fast it works, and what kind of customer it brings. Then it gives you a simple 90-day plan.
Start with the customer you want
Before choosing a channel, write down the job you want more of. "Anything with metal" leads to marketing that attracts a little of everything, including the jobs you'd rather not quote.
Answer five questions:
- Process: which of your processes has spare capacity and good margins? Laser cutting, press brake, 5-axis machining, robotic welding?
- Material and size: stainless, aluminum, plate, thin gauge, large weldments?
- Quantities: prototypes, short runs, or repeat production?
- Industries: who buys that work? Food equipment, transportation, agricultural machinery, construction, energy?
- Territory: how far will you deliver, and where do your best customers already sit?
The answer is your ideal job profile. Every channel below works better when you aim it at that profile instead of at "manufacturers".
The channels, compared
| Channel | Typical cost | Time to first job | Best for | Watch out for |
|---|---|---|---|---|
| Referrals and repeat work | Low | Unpredictable | Trust, good margins | You can't turn it up when you need it |
| Trade shows | High per show | 3 to 12 months | Meeting OEMs in one industry | Booth, travel and two days off the floor |
| Sales reps | Commission | 3 to 9 months | Regions you can't cover | Finding a rep who knows fabrication |
| Industrial directories | Low to medium | Variable | Being findable | Few direct requests in smaller markets |
| Manufacturing marketplaces | Platform takes a cut | Weeks | Filling gaps in capacity | Price pressure; the platform owns the customer |
| Cold outreach | Your time | 2 to 6 months | Targeting specific OEMs | Low reply rates without a sharp offer |
| SEO and content | Time or agency fees | 6 to 12 months | Long-term, low-cost requests | Slow to start |
| Google Ads | Ad spend plus management | 2 to 6 weeks | Predictable requests from buyers searching now | Wasted spend without negatives and tracking |
None of these is a silver bullet. Most growing shops end up with two or three working together: referrals for trust, one outbound channel for target accounts, and one inbound channel that brings buyers who are already looking.
Referrals: make them a system, not luck
Referrals bring the best customers, but most shops leave them to chance. A few habits turn them into a steady source:
- Ask at the right moment: right after a job ships on time and the customer says thanks. "Who else do you know who buys parts like this?"
- Stay in touch with past buyers. Purchasers and engineers change companies. A short note when someone moves can follow them to their new employer.
- Make it easy to refer you: a clear website that says exactly what you make, with your capacities and a quote form, gives the referrer something to forward.
Trade shows, reps and directories
Trade shows still work for shops that target one industry, because the buyers you want are all in one hall. They're expensive when you count the booth, travel and two days off the floor, so pick shows by the industry you want, not by habit, and book meetings before you go.
Manufacturers' reps can open regions you can't cover yourself, in exchange for commission. The good ones already sell complementary services (castings, finishing, electronics) to the buyers you want.
Industrial directories help you be findable, and they're worth a free or low-cost listing. In smaller markets like Quebec, though, they rarely produce a steady flow of requests on their own.
Marketplaces: useful, with limits
Online manufacturing marketplaces send you RFQs from buyers you'd never reach yourself, and they can fill a gap when the floor is quiet. The trade-off is that the platform usually sets or pressures the price, controls the relationship, and keeps the customer. Treat them as a capacity filler, not a growth strategy.
Cold outreach to target accounts
If you know exactly which OEMs you want (a list of 30 to 50 companies in your territory that buy your process), direct outreach works. The key is relevance: a short email or LinkedIn message to the right engineer or buyer, mentioning a specific capability that matters to their product, works far better than a generic capabilities brochure.
Inbound: be there when they search
The channels above all depend on you finding the buyer. Inbound flips it: the buyer finds you, at the moment they have a drawing and need a supplier. There are two ways to show up in Google.
SEO (organic search) means ranking your website for searches like "laser cutting service laval". It costs no money per click, but it takes time: typically 6 to 12 months of good content, a fast site and local signals like a complete Google Business Profile. For a local shop, the basics go a long way: one page per process, with materials, capacities, photos of real parts and a quote form.
Google Ads puts you at the top of those same searches within days, and you pay per click. Its strength is speed and control: you choose the processes, the territory and the budget, and you can measure exactly which searches turned into quote requests. Its weakness is that it costs money from day one, and without negative keywords and tracking, it can waste that money quickly. We explain the full setup in Google Ads for metal fabrication shops, and the budget math in How much does Google Ads cost?
Win more of the requests you get
More requests only help if you convert them. The shops that grow fastest do three things consistently:
- Answer within hours, not days. A buyer who sends a drawing to three shops often goes with the first credible answer. Set a target, such as 4 business hours, and hold the team to it.
- Qualify on the first call. Quantity, material, timeline, budget and whether there's repeat potential. Five minutes on the phone saves hours of estimating jobs you'll never win.
- Log every request. A shared sheet or CRM with the source, the value quoted and the outcome. After three months, you'll know which channel brings the jobs you actually want.
A 90-day plan
- Weeks 1 to 2: write your ideal job profile. Fix your website: one clear page per process, capacities, real photos, a quote form with drawing upload. Complete your Google Business Profile.
- Weeks 3 to 4: start a lead log. Follow up on every open quote. Ask your three best customers for a referral.
- Month 2: pick one inbound channel (usually Google Ads, for speed) and one outbound channel (a list of 30 target accounts). Launch both.
- Month 3: review the lead log. Which channel brought qualified requests? Which brought jobs? Put more behind what works and cut what doesn't.
If you want to run the inbound side yourself, the Fabricator's Google Ads Playbook has the step-by-step method, with planning tools and templates. If you'd rather have it built and managed for you, check whether your territory is open.
Calibre
Rather have it built for you?
Calibre sets up and runs Google Ads for one fabrication shop per process and region. Live in 14 days, and with at least $1,000 a month in ad spend, 10 quote requests a month by day 60 or we work for free.
Check if your territory is openWritten by
Frank Frenna
Co-founder of Calibre. Before Calibre, Frank ran the Google Ads campaign behind our Quebec sheet metal case study: 14 months, about 19 quote requests a month.
Gianluca Amato
Co-founder of Calibre, based in Quebec.