Google Ads for metal fabrication shops: the complete guide

Every day, an engineer somewhere in your region has a drawing, a quantity and a due date, and no supplier yet. They open Google and type something like "stainless laser cutting service" or "cnc machining prototype shop". Whoever shows up gets the drawing.

Google Ads is the most direct way to be that shop. You don't pay to interrupt anyone; you pay to answer people who are already looking for what you make. This guide explains how it works for metal fabricators specifically: laser cutting, sheet metal, CNC machining, welding and stamping shops. It covers how buyers search, how to structure the campaign, what to put on the page they land on, what to measure, and what to expect in the first 60 days.

Why Google Ads works for job shops

Most fabrication shops grow through referrals and repeat work. That's a good business, until a big customer slows down and the order book depends on who happens to think of you this week.

Search advertising fixes a specific problem: it puts you in front of buyers at the moment they need a supplier, not months before or after. Three things make it a good fit for job shops:

• The intent is in the search. "sheet metal fabrication quote" tells you more about the person than any demographic targeting could.
• You choose your territory. You can show ads only within 80 km of your shop, only in Quebec, or only in the industrial parks you deliver to.
• Every dollar is measurable. You can see which searches brought which quote requests, and what each request cost.

It's not for every shop. Google Ads works poorly if you can't answer a request within a few hours, if you're at capacity for the next six months, or if all your work comes through approved-vendor lists at three large OEMs. Fix those first.

How buyers search for a fabrication supplier

Buyer searches are built from a few blocks. Once you see them, keyword research becomes mechanical:

A search that combines a process with a supplier word, a location or a buying word is almost always a buyer. A process alone is mixed: some buyers, plus students, hobbyists and people shopping for a machine. A process with words like "jobs", "for sale" or "how to" is almost never a buyer.

That gives you four tiers to sort every keyword into:

1. Supplier searches: "laser cutting service laval", "custom sheet metal enclosure". Bid on these aggressively.
2. Process searches: "laser cutting", "aluminum welding". Bid carefully, with strong negatives and ads that pre-qualify.
3. Research searches: "laser cutting tolerances", "mig vs tig". Usually skip them in a paid campaign.
4. Wrong-audience searches: jobs, machines, DIY, unrelated meanings. Block them with negative keywords from day one.

Structure the account around what you make

The most common mistake in small fabrication accounts is too many campaigns, each with too little budget to learn anything. Google's automation and your own decisions both need volume.

A structure that works for most shops:

• One search campaign to start, unless you serve two very different markets. Split into separate campaigns only when a process or region has its own budget and clearly different economics.
• One ad group per process: laser cutting, bending and forming, welding, CNC machining. Each gets its own keywords, its own ads and, ideally, its own landing page section.
• Locations by radius or region, set to people in or regularly in your target area, not people "interested in" it. The default setting shows ads to people anywhere who search about your area.
• Both languages in Quebec. Many purchasers and engineers search in English even in francophone regions. Include French and English keywords and target both languages, with ads in each.
• Search only. Turn off the Display Network and search partners at launch. They add cheap clicks that rarely turn into quote requests.

Keywords, match types and negatives

Start with phrase and exact match on supplier and process searches. Broad match can work later, once the campaign records enough conversions for Google's bidding to learn from, but at launch it spends your money exploring.

A starting list for a laser-cutting ad group might look like this:

• "laser cutting service", "laser cutting company", "metal laser cutting"
• "stainless steel laser cutting", "aluminum laser cutting"
• "laser cutting montreal", "laser cutting near me"
• [custom laser cutting], [laser cutting quote]

Negative keywords matter as much as the keywords themselves. In fabrication accounts, a large share of wasted spend comes from four groups: job seekers (welder jobs, cnc machinist salary), hobbyists (laser cutting files, metal art), machine buyers (fiber laser for sale, used press brake) and unrelated meanings (laser hair removal, plasma donation). We've published 100 negative keywords to add before launch.

Ads that pre-qualify

Your ad has two jobs: attract the right buyer, and discourage the wrong one. Specific beats clever every time:

• Name the process and material in the headline: "Stainless & Aluminum Laser Cutting".
• Show capacity: sheet size, thickness range, tonnage, tolerances, quantities you like.
• Give a reason to choose you: lead times, in-house finishing, certifications you actually hold, bilingual service.
• Ask for the drawing: "Send your drawing, quote in 24 h". It sets the expectation that this is a B2B supplier.
• Filter politely: "Production and prototype runs for manufacturers" quietly tells the hobbyist this isn't for them.

Use every relevant asset: sitelinks to each process, callouts for certifications and lead times, structured snippets for materials, and a call asset during business hours.

The landing page: one job, quote requests

Sending ad traffic to your homepage is the second most expensive mistake. A homepage tries to talk to everyone: job applicants, existing customers, suppliers. A quote page talks to one person, the buyer with a drawing.

A good fabrication quote page has:

1. A headline that matches the search: "Laser cutting for manufacturers in Greater Montreal".
2. What you cut, bend, machine or weld, with materials, thicknesses and capacities in plain numbers.
3. Photos of real parts and your actual shop floor. Stock photos of sparks don't build trust.
4. Proof: industries served, certifications, years in business, a customer quote if you have permission.
5. A short form with file upload: name, company, email, phone, quantity, timeline, and a drawing upload that accepts PDF, DXF and STEP.
6. A response promise: "We reply within 4 business hours." Then keep it.
7. A phone number that's tappable on mobile.
8. French and English versions if you sell in Quebec.

On a focused page like this, it's common for 4 to 10% of ad clicks to become a call or form. On a generic homepage, 1 to 3% is more typical. That difference alone can cut your cost per request in half.

Measure requests, not clicks

If you can't see which clicks became quote requests, you can't tell good keywords from bad ones, and Google's bidding can't either. Before launch, set up conversion tracking for:

• Form submissions, counted on the thank-you page or a confirmed submission event.
• Calls from the ad itself, and calls from the website, using a Google forwarding number.
• Email clicks on the quote page, as a secondary action.

Then go one step further: log every lead in a shared sheet or your CRM, mark which ones were real requests and which jobs you won. Once a month, compare cost per qualified request and revenue won, not cost per click. That's how you optimize for jobs, not traffic.

Budget and bidding

You need two numbers before you set a budget: what a quote request is worth to your shop, and how many clicks a day Google needs to learn. As a rule of thumb, one campaign needs at least 8 to 10 clicks a day to produce data you can act on within a month. For industrial service searches in Canada, clicks often cost $3 to $8, so a single campaign in a regional market typically starts around $1,000 to $1,500 a month in ad spend.

We go through the full math, including what a lead is worth and what agencies charge, in How much does Google Ads cost for a fabrication shop?

For bidding, start with a strategy that doesn't need much data, such as Maximize clicks with a maximum cost per click, or manual bidding. Switch to Maximize conversions once the campaign records about 15 to 30 conversions a month and your tracking is verified.

What to expect in the first 60 days

• Week 1 to 2: the campaign goes live. Expect a lot of irrelevant search terms. Check them every other day and add negatives.
• Week 3 to 4: the first real quote requests arrive. Answer every one within a few business hours; a buyer who waits calls the next shop.
• Month 2: pause keywords that spend without producing requests, shift budget to the processes that do, and test new ad copy against the current best.
• Day 60: you should know your cost per quote request, which processes bring the best jobs, and whether to scale, hold or adjust.

Do it yourself or hire someone?

Running a fabrication campaign well takes a few hours a week: search terms, negatives, bids, ads, and the lead log. If you have that time and like the numbers, you can absolutely do it in-house. The Fabricator's Google Ads Playbook walks through every step, with planning tools and templates.

If you'd rather spend those hours on the shop floor, an agency that specializes in fabrication can get you live faster and avoid the expensive learning curve. Whatever you choose, insist on three things: conversion tracking on calls and forms, a weekly report in cost per request (not clicks), and access to your own Google Ads account.