I Cost My Shop $48,000 Buying the Wrong Equipment. Here Is My 7-Step Machine Purchase Checklist

I've run a small contract manufacturing shop handling metal fabrication orders for nine years. In that time, I have personally made and documented 12 significant equipment mistakes — roughly $48,000 in wasted budget. Some of that was scrap. More of it was idle machines and the quiet depreciation that follows a bad purchase.

When I first started buying equipment, I assumed the machine with the most impressive spec sheet was the right machine. It was not. What I learned is that a decision which looks obvious in a showroom can fall apart when real parts, real operators, and real monthly bills enter the picture.

This checklist is for other shop owners, operations managers, and purchasing people deciding whether to invest in a laser cutter, a CNC machining center, or injection molding capacity. It is also for anyone who keeps wondering whether the part they are machining should be outsourced instead. Seven steps. About two hours to work through. If you use it honestly, you will probably avoid the phrase I now hate: it will pay for itself.

Step 1: Audit the last 12 months, not the last 12 weeks

Open your invoicing or job log and sort what actually went through your shop in the last 12 months. How many hours of outside laser cutting did you pay for? How much CNC machining did you buy from other shops? What percentage of quoted jobs did you lose because you could not deliver the process?

The machine in your head should match the numbers on your spreadsheet. My suggested pass/fail check: if you cannot point to at least eight hours per week of external spend on that process, for two consecutive quarters, you do not have a capacity problem. You have a sales problem. A new machine will not fix that.

I learned this in 2017. A large customer told me they were insourcing their flat sheet cutting, and I assumed every other customer would follow. So I bought a fiber laser at auction. The wave never arrived. That machine logged 31 hours in its first six months, and I later sold it for $16,000 less than I had paid.

Step 2: Answer the 'is CNC machining expensive?' question with real numbers

Customers ask me all the time: is CNC machining expensive? My honest answer is that it depends on the denominator. CNC machining is expensive when you need two parts and you pay a shop's full setup cost. It stops looking expensive when you spread that setup across 500 parts.

Before you buy anything, get three real quotes for your most representative parts. Not the easy parts. The annoying ones with tight tolerances and difficult materials. Then build a basic total cost of ownership model for the machine you are considering:

• purchase price plus rigging and electrical work
• tooling, workholding, and first-year consumables
• CAD/CAM software and post-processing
• operator training time
• floor space, power, and scrap rate
• estimated hours per month of actual use, based on Step 1

Then compare the two paths over five years. Pass/fail: your in-house cost per hour needs to be at least 30 percent below the quoted external rate at your planned utilization. If it is not, outsource it and move on.

Step 3: Let geometry, material, and region pick the process

The old rule still mostly holds: flat sheet and plate profiles belong on a laser cutter; 3D blocks, deep holes, tapped features, and tight tolerances belong on CNC machining centers; high-volume plastic parts belong in injection molding. But the boundaries have shifted in the last five years. What was best practice in 2020 is not necessarily best practice in 2025. Fiber lasers cut thin sheet faster and cleaner than most job shops could have imagined a decade ago, and better CAM software has made CNC far less intimidating for smaller teams.

Look at your part family honestly. If you make enclosures that start life as flat aluminum sheet, a laser cutter is an easy yes. If your work is mostly round or block-shaped parts with milled pockets and threaded holes, a laser is the wrong tool. You are really choosing between a machining center and a good local CNC job shop.

Geography has a vote, too. I watched a shop in Oregon try to justify buying an injection molding press for more than a year. Then they spent one afternoon searching for injection molding Eugene and found two established molders within 30 minutes. The tooling cost still hurt, but the machine cost and the operator cost disappeared.

I used the same logic when I expanded my own shop in Tennessee. Searching for CNC machining Chattanooga TN surfaced a capable shop twenty minutes from my door. Any complex overflow part could go to them for less than the loan payment on a machining center. But what they could not do was quick-turn sheet metal work. That missing regional capability is what ultimately justified buying a laser cutter — not a hunch.

Step 4: Check the consumables ecosystem before you check the price

Do not ask what a machine costs until you know what it eats. Every machine is a system of consumables. A fiber laser needs nozzles, ceramic rings, protective windows, focus lenses, and eventually a replacement cutting head. A machining center needs end mills, toolholders, coolant, and filters. Injection molding needs resin and regular mold maintenance. Consumables determine uptime, and uptime determines whether the machine makes money.

Here is where I made one of my dumbest mistakes. I bought third-party laser nozzles to save about 40 cents each. They worked fine for about an hour. Then the beam quality degraded, the cut edge turned rough, and I scrapped roughly $700 worth of parts. The cheaper choice was only cheaper until it stopped holding tolerance.

Before our second laser purchase, I spent an afternoon looking up Bodor laser consumables. I checked the catalog, part numbers, and shipping times. It was the most boring afternoon of my decision process, and also the most useful. If replacement parts are hard to find, the machine is a paperweight waiting to happen.

Pass/fail: you should be able to order a consumable kit and receive it within three days at your actual shop address.

Step 5: Map the support network in your region

A machine is only as useful as the technician who can reach it. Before buying, find the dealer, the service network, and — this matters more than people admit — other shops running the same brand nearby. Other owners will tell you things salespeople cannot or will not: which component fails first, which consumable is overpriced, which software update broke a post processor.

When a colleague in the UK asked me to help her evaluate a machine, the strongest signal was the installed base. Searching for Bodor laser cutters UK surfaced multiple dealers and plenty of shop owners running them. If the same search had come up empty, I would have told her to walk.

Pass/fail: you should be able to name a service contact, get a response within one business day, and find at least two nearby shops that own the same machine and are willing to talk.

Step 6: Test with your parts, your material, your operator

Demo parts lie. The vendor's sample was cut on perfectly tuned settings, with fresh consumables, by someone who has run that specific machine for years. When the sales engineer made a clean 5 mm mild steel part look effortless, I was impressed. Then we loaded our actual 1.2 mm stainless sheet with small internal features. The speed dropped by about 40 percent, and the edge quality was not good enough without a second pass.

Take your own CAD files, bring your own material, and ask the person who would actually run the machine to be there. Let them operate it. If the machine only performs when the vendor's expert is in control, that is not a machine you should buy.

For thermal cutting, ask for cut-quality measurements, not just a good-looking edge. The ISO 9013 standard gives both sides a neutral way to talk about edge roughness, dross, and perpendicularity. If the vendor cannot match your required quality class on your material, no amount of advertised speed matters.

Step 7: Put the purchase order in a drawer for 30 days

If a machine still makes sense after every step above, the final step is to wait 30 days. Not for a better price. For your own brain.

I have never once needed a new machine so urgently that a 30-day wait hurt. But I have made at least three purchases where urgency was manufactured, and all three were bad. Give the spreadsheet time to catch up with your excitement. Re-read your Step 1 numbers. Phone one more existing owner. If the decision survives the pause, buy it with confidence.

What I will not do again

To make this checklist easier to use, here are the red flags I now treat as automatic stops:

• Any quote that is only valid today. Machines are not airline tickets.
• A vendor who cannot give consumable part numbers during the sales process.
• A demo that uses different material than the material you actually cut.
• The phrase this machine will pay for itself coming out of my own mouth. When I hear myself say it, I know I am selling instead of analyzing.

In nine years, I have made almost every equipment mistake in this article. The good news is that the last two purchasing decisions — both laser-related — have been the best investments we have made as a shop. Not because I am smarter now. Because I finally have a process. This seven-step checklist is that process. Steal it.