3D printing needs no tooling and is fastest for a few parts; injection molding needs a mold up front but then drops the cost per part sharply. The rule: print for prototypes and tiny quantities, mold once you need many identical parts — there's a crossover quantity where molding becomes the cheaper total.
| 3D printing (additive) | Injection molding | |
|---|---|---|
| Tooling | None — print straight from CAD | A mold must be built first (up-front cost) |
| First part | In days; change the design freely between prints | After the tool is built; design changes mean modifying the tool |
| Cost per part | Flat — each part costs about the same as the last | Tooling amortized over the quantity, so very low per part at volume |
| Best quantity | One-offs to low volume | Many identical parts, at scale |
| Strength | Often anisotropic — weakest between layers (the Z direction); worst on FDM, less so on SLS/MJF | Far more uniform and consistent batch to batch (though weld lines and fiber fill still add some anisotropy) |
| Materials / finish | Process-specific (FDM: PLA/PETG; SLS/MJF: nylon); often visible layer lines | Wide range of production resins (ABS, PP, PC, glass-filled nylon…); consistent finish |
The economics flip on one idea. Printing has near-zero setup but a flat per-part cost; molding has a high up-front tooling cost but a much lower per-part cost. Plot both and the lines cross at a break-even quantity — below it, printing is cheaper overall; above it, molding wins and keeps winning. The crossover depends on part size, cavity count, tool class (aluminum bridge vs hardened-steel production tool), and geometry — but for a product sold in any real quantity it is usually reached at modest numbers, which is why most mass-produced plastic parts are molded. In symbols: break-even ≈ tooling ÷ (print per-part − molded per-part). (See what drives mold cost.)
Most products use both, in order:
The key reason not to skip straight from a print to full production: a printed part checks shape, but it doesn't behave like the molded part — different strength (layer anisotropy), feel, and finish — so a real molded sample belongs before you commit to volume.
And a printed design often has to be redesigned for molding. Printing ignores the rules molding lives by — draft angles, undercuts, uniform wall thickness — so a shape that prints fine may need DFM changes before it can be tooled. Catch that early, not after you've committed to a mold. (And printing can stay the right answer even at volume for complex, highly custom, or genuinely low-run geometry.)
The honest way to choose is to find the volume where the two cross over on cost. 3D printing has almost no setup cost but a high, flat cost per part, so ten parts and a thousand parts cost roughly the same each. Injection moulding has a large one-time tooling cost but a very low cost per part, so the tool's cost spreads thinner across every part you make. Below the crossover — a handful to a few hundred parts, depending on size and resin — printing is cheaper because you never pay for a tool. Above it, moulding wins and keeps winning, because each extra part costs only material and machine time. The mistake is committing to a tool for a volume that never reaches the crossover, or printing at a volume where a tool would have paid for itself. Our cost calculator and mould-cost estimator put rough numbers on where your part crosses over.
Cost is only half the decision; the parts themselves differ. A printed part is built up in layers, so it carries layer lines, anisotropic strength (weaker across the layers than along them), and material properties that rarely match a moulded engineering resin. A moulded part is a solid, far more uniform piece in the actual production material, with the surface finish, tolerances and strength the end product needs. That means a printed prototype can confirm shape, fit and ergonomics but can mislead on strength, sealing, clarity or how the real resin flows and warps — exactly the things a moulded part reveals. Treating a printed part as a proxy for the moulded one is fine for form and fit and risky for function, which is why bridge tooling exists to make real moulded parts before the production tool is cut.
The two are not rivals so much as stages, and the smartest programmes use both. Printing earns its place early — iterating the design cheaply, checking fit and ergonomics in hand, making a few units for a pitch or a user test — because changing a printed design costs nothing but a reslice. Once the design is proven and the volume justifies it, moulding takes over for production economics and real-material parts. Printing can even stay useful after moulding starts, for jigs, fixtures, low-volume variants and spares that would never justify their own tool. The design sequence that wastes the least money is to print while the design is still moving, commit to tooling only when the design is frozen and the volume is real, and keep a printer around for the long tail — rather than forcing one process to do the other's job.
For a given part the decision comes down to a few honest questions: how many will you make over its life, is the design frozen or still changing, does the part need production-grade material and finish, and how fast do you need the first real units. High volume, frozen design and a need for real material point to moulding; low or uncertain volume, a moving design or a need for parts this week point to printing, often with bridge tooling as the middle step that serves a first market in real material while the production tool is built. We will tell you honestly which side of the line your part sits on, and when the answer is "print now, tool later," we say so rather than selling a tool you do not yet need.
Which should I use? Print for prototypes/tiny quantities and design freedom; mold for many identical parts (lower per-part cost past the crossover quantity).
Is a printed part as strong? Usually not — printing is anisotropic (weakest across the layers, in Z); a molded part is far more uniform, though weld lines and fiber orientation add some anisotropy.
When does molding get cheaper? At the break-even quantity where its high tooling + low per-part cost beats printing's flat per-part cost — modest quantities for most products.
Can I use both? Yes, in sequence: print to learn, bridge to launch, mold to scale.
3D printing's limits? Slow/costly per part at volume, narrower materials/finish, layer-line strength limits.
JCH Design & Manufacturing can tell you honestly whether you're still at the printing stage, ready for tooling, or somewhere in between — and what each path costs in time and risk. Send your part and expected quantity; NDA first.
Book a 30-minute callTrying to decide whether your quantity is still a 3D-printing job or time for a mold? Ask our AI assistant. It answers from this website, and Jack follows up if you want.
AI assistant. Answers can be wrong; nothing it says is a quote or a promise.