The 9-Step Laser Job Checklist I Built After Burning $7,000 in Mistakes

Who this checklist is for (and what it covers)

If you're running a laser machine — a 40-watt diode engraver, a cheap laser cutter you grabbed off a marketplace listing, or a fiber marker sitting on the shop floor — this is for you. I've been handling laser marking and engraving orders for seven years. Before that, I spent two years on the medical side, working with Candela-type platforms (GentleMax Pro Plus, Nordlys) at an aesthetic clinic.

Laser is laser. The mistakes I've made in both worlds add up to roughly $7,000 — maybe closer to eight, I stopped counting after 2023 — and too many late nights fixing things that shouldn't have needed fixing.

There are nine steps. I run through them in the same order on every job. It's muscle memory now.

Step 1: Verify the material, don't guess

"It looks like anodized aluminum" has cost me more money than any machine failure. Aluminum gets sold as anodized when it's had a coating so thin you could sneeze it off. Laser hits it, and you get either a faint ghost mark or a burn that goes straight through.

In 2019, I ran an 80-piece order on what the client swore was 6061 anodized. It wasn't. We got thirty pieces in before noticing the marks were coming out uneven and dull. That job cost $890 in redo plus a week of delay. I still kick myself for not running a 30-second check on the first piece. Now, before I buy aluminum for any job, I check surface resistance with a multimeter and do a cross-hatch test on a scrap corner. Real anodizing scratches white, never bright metal.

Checkpoint: cut a test piece off the edge stock before you touch the real job. This one habit has saved me at least four times since.

Step 2: Match the laser type to the material

The step most people skip. Not every laser does every job. Fiber, CO2, diode, UV, MOPA — different wavelengths absorb into different materials in completely different ways.

Same logic applies on the medical side. People ask me "laser Candela or laser diodo?" all the time, and the honest answer is: it depends what you're treating. A Candela pulsed-dye platform and a 40-watt diode engraver on my shop bench are not the same species. Different wavelength, different pulse profile, different job entirely. Anyone comparing them hasn't figured out what they actually need yet.

Quick mapping that keeps me out of trouble:

• Anodized aluminum marking → fiber or MOPA, no exceptions
• Acrylic cutting or engraving → CO2
• Light woods, leather → diode is fine
• Stainless color marking → MOPA
• Clear plastics or precision medical parts → UV, but that's a different budget tier

Back in 2022, I tried a diode on a job that needed fiber. Forty pieces later the marks looked like pencil smudges. $1,100 gone. That's when I built the mapping card and hung it by the door.

Step 3: Run a scrap test — always

Power, speed, frequency, focus — put them all on scrap first. Doesn't matter if you've run the job a hundred times. Material batches change, machines drift, your memory lies to you.

We keep a "scrap bin" next to every machine for exactly this. Fifty cents of waste material has bailed me out of hundreds of dollars in rework.

The March 2021 rush job changed how I think about skipping the scrap test. A client needed 200 pieces in two days, and I decided the test step was "optional this time." Fifteen minutes saved, three days lost to a redo. Not worth it, ever.

Step 4: Check the file before the machine

File problems cost more than machine problems. 300 DPI is the industry standard for commercial print and marking, but I still get files at 72 DPI from people who sized them in a PowerPoint slide.

Print size (inches) = pixel dimensions ÷ DPI. A 3000 × 2000 pixel image at 300 DPI maxes out at 10 × 6.67 inches. Anything bigger and you're stretching pixels.

For brand colors, Pantone tolerance matters. Delta E under 2 is brand-grade; 2–4 is noticeable to trained eyes; above 4 most people see it. I keep a Pantone Color Bridge next to my monitor because screens lie, and I've had a client reject a run of marks based on a color shift I couldn't see on my calibrated display.

And Pantone colors don't have exact CMYK equivalents. Pantone 286 C converts to roughly CMYK 100/66/0/2, but the actual printed result depends on substrate and calibration. Don't trust "looks close on screen" as your standard.

Step 5: Anodized aluminum — don't crank the power

Laser engraved anodized aluminum works by stripping the dye layer to reveal the pale oxide underneath. It's not burning — it's revealing. Once you understand that, the settings make sense.

For a 20W fiber, 30–40% power at 300–500 mm/s gives clean white marks. Crank it higher and you'll get bubbling, darkening, or bleed through to the base metal — which nobody can polish out.

Anodizing thickness varies, even within a single batch of the same sheet. In 2023, three sheets out of a "same supplier, same spec" run came in noticeably yellow, and the marks came out a full shade off with identical settings. $620 wasted. Now I pull three test pieces from different spots on every new sheet lot. Non-negotiable.

Step 6: Acrylic — speed and focus are everything

Most of what I see written about how to laser engrave acrylic misses the actual sticking points:

• Cut with CO2, not diode. Diode wavelengths pass through clear acrylic without cutting.
• Engrave clear acrylic with a black card underneath. Otherwise the beam passes straight through and leaves nothing.
• Too fast = white haze. Too slow = melted yellow edge.
• Focus off by 1 mm = a visible quality drop on the edge. Refocus every time you change thickness.

A shop I helped last year couldn't cut 5 mm acrylic on their cheap laser cutter. Machine said 60W, the actual output was closer to 40W, and a dirty lens was eating another 30% on top. A lens clean and a 15% power bump fixed it. Rule of thumb: when a cut fails, clean the lens before you touch the power setting.

Step 7: Budget laser cutters — the honest version

I bought a $400 "cheap laser cutter" in my second year. It works, but only if you accept the terms:

• Rated power is optimistic. A "60W" diode or budget CO2 often delivers 40W real output.
• Mirrors and lenses need routine alignment and cleaning, or output drops week over week.
• Hobby exhaust systems are decorative. Acrylic fumes need a real external vent.
• Included software is often locked to proprietary formats.

This doesn't mean don't buy one. Small batches, soft materials, no deadline pressure — it'll do the job. Expect it to run a small production shop and you'll be buying a second machine within six months. I did. That's how I know.

Step 8: Pre-run check before you press go

Sixty seconds before every run:

1. Material tested on scrap?
2. Focus verified?
3. Exhaust on?
4. File 1:1 scale at 300 DPI minimum?
5. Workpiece clamped or weighted down?
6. Power settings match the test piece?

Every one of those six lines exists because I skipped it once. People told me "always test before running the real job." I didn't listen. That one shortcut turned a 15-minute fix into a 3-day problem. Now the list is on a sticky note next to the emergency stop button, where I can't pretend not to see it.

Step 9: Post-run inspection

Don't ship straight off the bed. Pull three pieces. Check edge quality, contrast, and whether anything got skipped.

Fresh off the machine, parts are hot — and if it's dyed anodized, wiping with alcohol too soon can smear the dye layer. Let it cool, then wipe with a lint-free cloth.

Things that bite people (and bit me)

1. Don't trust displayed power. Ambient temperature, tube age, and material batch all change what actually lands on the workpiece. "50% on the screen" is not "50% output."

2. Cheap material hurts more than a cheap machine. Machines you can calibrate. Materials you can only re-buy.

3. If a client is weighing a Candela-class laser device against a generic diode aesthetic system, the use case matters more than the spec sheet. I've seen people buy the wrong machine and come back six months later wanting to trade in. There's no single best laser — there's a right one for the job in front of you.

4. Telling a client "this isn't the one for you" builds more trust than upselling. About 15% of our acrylic inquiries — maybe 18%, it varies by month — end up being referred to another supplier because our machines genuinely can't cover what they need. Roughly a third of those referrals come back within a year with a job that does fit.

This list isn't finished. Every job adds a note. Yours will look different than mine, because your mistakes will be different than mine. But the moment you start writing them down, you're already ahead of about 80% of the people running lasers right now.