The crossover between 3D printing and injection molding falls between 300 and 3,000 units for most consumer parts, and part size is what moves it. A 5-gram clip that prints for $4.50 and molds for $0.09 in a $12,000 tool crosses over at about 2,700 units. A 600-gram tote that prints for $180 and molds for $2.40 in a $55,000 tool crosses over at about 310 units. Bigger parts push the crossover down, because printing cost scales with volume of material while molding cost per part barely moves.
That single relationship decides most of these arguments. The rest is material properties, tolerance, and how many design changes you still expect to make.
The Breakeven Formula
The math is one line:
Breakeven units = tooling cost ÷ (printed cost per part − molded cost per part)
Everything else is figuring out the three inputs honestly.
Worked case. A 42-gram ABS housing molds at $0.62 per part in a $26,000 two-cavity steel tool. The same geometry in MJF nylon at a service bureau, at quantities of 100, prices around $22 per part.
Breakeven = 26,000 ÷ (22.00 − 0.62) = 1,216 units.
Under 1,216 units, printing is cheaper. Over it, molding is. But notice how sensitive that number is to the tooling figure. Swap the $26,000 production tool for an $8,000 aluminum bridge tool with a $0.85 piece price and the breakeven drops to 378 units. Tooling choice, not printing technology, is the lever that moves the answer.
How Part Size Changes the Answer
Printing charges by material volume and machine time. A part twice as big costs roughly twice as much to print. Molding charges by machine time and resin, and a part twice as big adds maybe 40% to piece price while adding 30% to 60% to tool cost. The two curves diverge.
| Part | Printed cost (qty 100) | Molded piece price | Tool cost | Breakeven |
|---|---|---|---|---|
| 5 g clip | $4.50 | $0.09 | $12,000 | 2,721 units |
| 42 g housing | $22.00 | $0.62 | $26,000 | 1,216 units |
| 180 g enclosure | $68.00 | $1.15 | $38,000 | 568 units |
| 600 g tote | $180.00 | $2.40 | $55,000 | 310 units |
The practical read: if your part is small and cheap to print, you can print farther into your launch than you think. If your part is large, tooling pays back fast and printing is an expensive way to stall.
Material Properties and Tolerances Are Not Equivalent
Cost is the argument people have. Material behavior is the one that ends projects.
Anisotropy. Fused deposition parts are weaker in the Z axis, across the layer lines, than in X and Y. Depending on material and print settings, Z-axis tensile strength runs 30% to 70% of in-plane strength. A molded part is isotropic. If your part sees load in a direction the printer laid down as a layer boundary, printed test results will not predict molded behavior, and molded results will not predict printed behavior either.
Powder-bed nylon versus molded nylon. MJF and SLS PA12 parts reach roughly 80% to 95% of injection-molded PA12 tensile strength, with lower elongation at break. They are close enough for functional testing, not close enough for a fatigue-critical part.
Stereolithography resins. SLA and DLP parts have excellent surface finish and dimensional accuracy but most standard resins are brittle, with elongation at break under 10%, and they yellow and lose strength under UV. They are the wrong choice for a snap-fit that gets cycled or for anything living outdoors.
Living hinges. Polypropylene living hinges survive hundreds of thousands of cycles when molded, because the molding process orients the polymer chains across the hinge. No printing process reproduces that. If your design has a living hinge, printed prototypes will fail and tell you nothing.
Surface and porosity. Powder-bed parts have a matte, slightly porous surface that holds dirt and does not seal against liquids without post-processing. If the part needs to be watertight or food-contact compliant, molding is the path.
Molding also holds tighter tolerances, and holds them across the run.
| Process | Typical tolerance | Repeatability across a run |
|---|---|---|
| Injection molding | ±0.05 to ±0.13 mm | High after process is dialed in |
| CNC machining | ±0.025 to ±0.08 mm | High |
| SLA / DLP | ±0.1 mm or ±0.2% | Moderate, varies by build position |
| MJF / SLS | ±0.2 mm or ±0.3% | Moderate, varies by packing density |
| FDM | ±0.3 mm or ±0.5% | Lower, varies by machine and material |
Printed part dimensions vary with where the part sat in the build chamber. Two identical files printed in the same job can differ by 0.2 mm depending on thermal history. For a housing that has to snap into a mating part, that variation shows up as parts that fit and parts that do not, from the same build.
Speed and Iteration Cost
Printing wins on change cost, and it is not close.
A design revision on a printed part costs a file edit and a new build: 1 to 5 days and the price of one part. A revision on a molded part after steel is cut costs $800 to $12,000 in tooling rework and 2 to 6 weeks of schedule. Steel safe changes, where you remove material from the cavity to add material to the part, are the cheap direction. Adding steel back, which means making the part thinner or smaller, requires welding and remachining or a new insert.
This is the real reason to stay on printing longer than the cost curve suggests. If you have any doubt left about the geometry, the fit, or the user interaction, the option value of cheap changes exceeds the piece price penalty. Sorting that out is exactly what prototype iteration is for, and it belongs before tooling, not after.
Where Each One Wins
Print when:
- Annual volume is under the breakeven you calculated, not the one you hope for
- The design is still changing, or user testing is not finished
- You need parts in under two weeks
- You need 10 to 500 units for a sales sample program, a crowdfunding fulfillment, or a regulatory submission
- Geometry is impossible to mold: internal lattices, fully enclosed voids, conformal channels
- You need multiple variants of the same part in small quantities
Mold when:
- Volume clears the breakeven with margin
- The part needs molded-only properties: living hinges, overmolds, clear optical surfaces, sealed housings
- Cosmetic finish matters and you do not want to pay for hand finishing on every unit
- The design is frozen and validated
- You need dimensional consistency across thousands of units
- Cost per part has to fall under $2
The Middle Path: Bridge Tooling
The choice is not binary. Bridge tooling is a soft aluminum mold, usually single cavity, built in 2 to 4 weeks for $3,000 to $9,000, good for 1,000 to 10,000 shots. It gets you molded material properties, molded surface finish, and molded tolerances at a fraction of production tooling cost.
The trade is tool life and cycle time. Aluminum cools faster than steel, which helps cycle, but it wears, galls on undercuts, and will not survive glass-filled resin. Run 5,000 units off a bridge tool, sell them, learn what the market wants, then cut production steel from a design informed by real demand. That sequence, along with urethane casting and low-cavitation soft tools, is covered in the low-volume manufacturing options.
A Decision Sequence That Works
- Estimate annual volume for year one, then cut it in half. Inventors overestimate year one by a wide margin, and the cost of tooling too early is worse than the cost of tooling too late.
- Get a real molding quote with tooling and piece price broken out. Not an estimate, a quote. The RFQ package a molder needs is short.
- Get a printed quote for the same file at 100 and at 500 units.
- Run the breakeven formula with the real numbers.
- Ask whether the design is frozen. If it is not, add 12 to 18 months of printing to the plan, or use a bridge tool.
- Check whether any molded-only property is load-bearing in your design. If yes, the crossover argument is over and you need at least a bridge tool for validation.
One thing printing does not fix: it does not remove the need for design for manufacturability work. A part designed with no draft, non-uniform walls, and eleven undercuts will print fine and quote terribly when you eventually mold it. Teams that print through their whole development cycle and then hand the file to a molder get a tooling quote 40% to 80% above what a moldable version of the same design would cost, plus 4 to 8 weeks of redesign.
Design for the process you intend to end on. Print the parts, but draft the walls.
Protecting the Design Before It Leaves Your Desk
Both paths mean sending 3D files to a vendor. A printed part from a service bureau and a molded part from a contract manufacturer both start with your geometry sitting on someone else’s server. File a provisional first, or at minimum run a $399 patent search against the USPTO patent records so you know what you are protecting and whether it is clear. USPTO’s overview of the patent process lays out where a provisional fits in the sequence and what a filing date buys you.
Then get the manufacturing decision right, because it sets the unit economics you will live with for years. Enhance’s manufacturing sourcing work covers exactly this evaluation, and the tooling piece is broken down further in what injection molding tooling costs.
FAQ
At what quantity does injection molding become cheaper than 3D printing?
For a small consumer part, somewhere between 800 and 3,000 units. For a large part, as low as 250 to 500. Run the numbers with your own tooling quote, because the tool price is the dominant term and it varies by a factor of five across part geometries.
Can I use 3D printed parts to sell a real product?
Yes, and plenty of low-volume hardware businesses do. The constraints are unit cost, surface finish, and consistency. Printed parts also carry post-processing labor of 3 to 20 minutes each for support removal, bead blasting, and dyeing, which is real cost that quotes sometimes bury.
Is a 3D printed prototype good enough to test my design?
For fit, ergonomics, and user interaction, yes. For structural performance, only if you account for anisotropy and material differences. For living hinges, overmolds, and optical clarity, no. Match the prototype method to the question you are asking rather than to the machine you have access to.
What is bridge tooling and when does it make sense?
A soft aluminum mold, $3,000 to $9,000, 2 to 4 week lead time, good for 1,000 to 10,000 shots. It makes sense when you need molded properties before you can justify production steel, when you want market data before committing $30,000 to tooling, or when your first order is 3,000 units and your forecast is a guess.
How do these choices affect my total launch budget?
Tooling is usually the single largest line item in a hardware launch after engineering. A $26,000 mold is often 30% to 50% of the total cash needed to reach first shipment. The complete inventor cost breakdown puts that number next to patent, design, and marketing costs so you can see the whole picture before committing to either path.