I opened the quote from an injection mold shop and almost closed my laptop. $22,000 for a bracket the size of a smartphone. The part itself was $0.18 per unit, but the tooling cost made it feel like buying a car to avoid walking.
My first thought was: a 3D printer can do that for under $20 in material. That's where a lot of procurement mistakes start, including mine. Let me rephrase that: I almost compared a raw material cost to a fully engineered manufacturing process. They're not the same thing.
The Surface Problem: A Tooling Quote That Feels Like a Car Down Payment
If you're in procurement, you've felt this. You send a part file to six plastic injection molding tooling manufacturers. Three quote you between $12,000 and $28,000. One says 'can't help.' Another says 'call us.' You have no idea which number is real.
Over the past six years, I've tracked roughly $180,000 in manufacturing spending in a simple cost-tracking spreadsheet. In Q4 2024, I asked 12 plastic injection molding tooling manufacturers to quote a simple 90 mm × 60 mm housing. Same part, same PDF, same annual volume. The quotes ranged from $9,400 to $26,700. Honestly, I'm not sure why some shops were double the others. My best guess is that one shop was busy and didn't want the work, and another was hungry and quoted lean. That's the part nobody puts in the brochure.
What most people don't realize is that a tooling quote isn't a single product. It's engineering, mold base, cavity machining, electrode work, heat treatment, polishing, assembly, and usually a few trial shots that get scrapped. When I unpack the quotes, the scope varied more than the price:
- two quotes included first articles; the rest didn't
- one included mold design; another treated it as a 'cost plus' line item
- none of the quotes said the same thing about spare parts or wear items
Asking why a mold is so expensive is like asking why a factory costs so much. The answer depends on what's inside. At least, that's been my experience with roughly 30 tooling projects in industrial equipment. If you're buying high-volume consumer goods, your numbers and your vendors will be different. I can't speak to that segment.
The Deeper Problem: Capability Is Not the Same as Economic Fit
Here's the part that took me years to understand: a manufacturing method's capability has almost nothing to do with whether you should use it.
A client emailed me last month: 'Can a laser welder weld aluminum?' The short answer is yes. A pulsed fiber laser can weld aluminum, and per The Welding Institute's materials guidance, the 1 μm wavelength is absorbed far better than older CO2 lasers. The long answer is: how many aluminum welds do you actually need next year? If the answer is three, buying or even renting a laser welder is the wrong move. The capability is real. The economics aren't.
Same logic applies to machining. A vertical machining center (VMC) can easily make a one-off bracket. But using a $150,000 VMC to produce one prototype is like buying a frozen-margarita machine for one drink: it works, it's capable, and it's the wrong economic decision. 'Can' is a technical question. 'Should' is a total-cost question.
Injection molding can make millions of identical parts. That's its power. It can also cost you $15,000 before the first part ships. If you need 200 parts, the mold isn't the real problem. The assumption that you need a mold at all is the problem.
The Real Cost: What a Wrong Process Decision Does to Your Budget
I still kick myself for the time we approved a steel mold for a part with insufficient draft. The supplier flagged it, but we didn't understand what that meant. The first trial parts stuck in the cavity. We paid for steel welding, re-cutting, polishing, and lost eleven days of schedule. A $14 3D-printed prototype would have shown the geometry problem the day we held it in our hands.
That's the hidden cost that never appears on a quote: the cost of learning after you've spent the tooling money.
Let's put some numbers around the process decision. These are from my own cost system, not an industry benchmark, because your numbers will differ:
Simple mold: $15,000. Fully loaded molded part: $2. Fully loaded FDM part on a professional Ultimaker-class printer: $15. Break-even volume = $15,000 ÷ ($15 - $2) ≈ 1,154 parts.
At 100 parts, molding is $15,200 and printing is $1,500. At 1,000 parts, molding is $17,000 and printing is $15,000. At 10,000 parts, the mold wins badly: $35,000 versus $150,000. This isn't a story about 3D printing replacing injection molding. It's a story about matching the process to the volume.
Remember, fully loaded should include machine hours, operator time, post-processing, and scrap. If you forget scrap, you'll convince yourself that the $15 number is too high. It's not. It's just honest.
The Honest Solution: Let the Break-Even Number Decide
I recommend 3D printing when you need fewer than roughly 1,000 to 1,200 parts, can tolerate an FDM surface finish, and don't need exotic material properties. That's a wide range, but it fits a lot of spare parts, jigs, fixtures, and bridge production runs. If your volumes are 10,000+, talk to plastic injection molding tooling manufacturers and pay the tooling cost. The mold is the better investment.
If you're searching for 'Ultimaker S8 3D printer specifications' to help make that decision, I'll save you some time: there is no official Ultimaker S8 spec sheet as of January 2025. The current S7 model lists a build volume of 330 × 240 × 300 mm on Ultimaker's official product page, which is the benchmark I use for parts up to that size. The next model, when it arrives, won't change the break-even formula. It will just move the line slightly.
Ultimaker 3D printer filament matters more than the machine model. Tough PLA is fine for fit checks and many functional jigs. If you need chemical resistance or higher heat tolerance, look at PETG, PP, or a polycarbonate blend. If you need ABS-like strength, you need the right print chamber, and you need to run the drying and printing settings exactly. Ultimaker's material compatibility list tells you what works with the machine. It doesn't tell you whether 3D printing is the right process—that's still your call.
Why I Still Pay for Injection Molds
Let me say this clearly: 3D printing does not replace injection molding. Injection molding is faster per part, has better material consistency, and the unit cost drops with volume. You will not 3D print 50,000 brackets in a week. I'm not recommending that.
I'm recommending that you stop treating a $20,000 mold quote as the only option. The honest calculation is the fully loaded cost of each path, including your time, your scrap rate, and your schedule. If your volume is above the break-even, pay the mold maker. If it isn't, buy the filament and get to work.
Oh, and the answer to 'can a laser welder weld aluminum' is still yes. But the correct next question is: should I buy a laser welder, or outsource this one job to someone with experience? Most of the time, you already know the answer.
