3D Printing vs Injection Moulding: When Should You Actually Switch?

Every hardware business reaches this decision. You've a product that works, demand that's growing, and a quote from a tool maker sitting in your inbox with a number on it that makes you uncomfortable.

Do you commit to the tool, or keep printing?

The usual answer, "print below a thousand units, mould above", is roughly right and completely useless, because the actual crossover depends on things that vary enormously between projects. Here's a framework that gives you a real answer for your specific part.

Two completely different cost structures

The decision isn't really about which process is better. It's about which cost curve you want to be on.

Injection moulding front-loads everything. You pay for a steel tool, in India, anywhere from ₹1.5 lakh for something simple to ₹15 lakh or more for a complex multi-cavity mould, and wait four to twelve weeks for it. After that, each part costs very little, sometimes only tens of rupees. The curve is a huge step followed by an almost flat line.

3D printing has no step at all. There's no tool, so the first part costs about the same as the hundredth. The curve is a straight line from the origin, with a modest downward slope from volume discounts (up to 18% at 100+ units).

Draw both curves and they cross somewhere. Everything below is printing territory; everything above is moulding territory. The whole question is where that crossing point sits for your part.

Where the crossover actually falls

It moves a lot, driven by three things.

Part size. A small part might print for ₹150 and mould for ₹15, so the tool pays back over a few thousand units. A large part might print for ₹2,000 and mould for ₹200, the per-unit gap is bigger in absolute terms, so payback comes sooner in unit count but the tool is also more expensive. Larger parts generally favour moulding earlier.

Part complexity. Undercuts, side actions and complex geometry make tools dramatically more expensive, each side action can add lakhs. Printing doesn't care about undercuts at all. The more complex the part, the further out the crossover moves, sometimes far enough that moulding never makes sense.

Material. Commodity moulding resins are cheap. Filament isn't. A PLA part has a relatively small per-unit gap; a PA-CF part has a large one.

As a rough guide for a moderately complex part of moderate size: the crossover typically lands somewhere between 1,000 and 5,000 units. But treat that as a starting hypothesis to check, not an answer.

The three questions that actually decide it

Unit economics are only part of the picture, and often not the deciding part.

1. Is the design actually final?

This is the big one, and it's where most businesses get hurt.

A steel tool encodes your design in metal. Minor changes might be possible by modifying the tool. Significant changes mean a new tool and the full cost again.

If there's any real chance the design will change, because you're still getting customer feedback, because a component might be superseded, because you haven't finished testing, then tooling early is a bet you may lose entirely. Printing carries zero design risk. Revision seven costs the same as revision one.

Rule of thumb: don't cut steel until you've shipped a few hundred units of the current design and nobody has asked you to change it.

2. How confident are you in the demand?

A tool is capital committed against a forecast. If you sell 500 units of a product you tooled for 10,000, you've converted cash into a steel object that's worth almost nothing.

Printing lets demand pay for itself. You make what you can sell, and if the product doesn't take off you've lost nothing beyond the parts you actually made. When demand is proven and repeating, the tool becomes an obvious investment rather than a bet.

3. Can you wait?

Tooling lead times run four to twelve weeks before the first part exists, plus trial-and-adjust iterations. Printing delivers in days.

If you've a launch date, a customer commitment or a competitor moving, printing bridges the gap. Plenty of businesses ship the first several hundred units printed while the tool is being cut, then switch. That isn't a compromise, it's the correct sequencing.

What moulding does better

Being fair about this matters, because the wrong answer is expensive in both directions.

Unit cost at volume, not close, once the tool is paid for. Surface finish, moulded parts come out glossy and uniform with no layer lines. Isotropic strength, moulded parts are equally strong in all directions; printed parts aren't. Material range, vastly wider selection of resins, grades, fillers and additives. Cycle time, seconds per part against hours. Consistency at scale, a well-tuned mould produces ten thousand near-identical parts.

If you're making a simple part in commodity plastic in genuine volume, moulding is correct and you should tool up.

What printing does better

No tooling cost or lead time. The entire step in the cost curve disappears.

Design freedom. Undercuts, internal channels, lattices and consolidated assemblies are free. A nine-piece welded assembly can become one printed part, which often means the printed version is cheaper in total even when the individual part isn't, because you've eliminated eight parts and the labour of assembling them.

Zero cost to iterate. Revision costs nothing.

Economic at any quantity. One part is fine. So is three hundred.

Mass customisation. Every part can be different at no extra cost. Serial numbers, customer names, size variants. A mould can't do this at all.

No minimum order. No obligation to buy 5,000 units of anything.

The bridge production strategy

The framing of "printing versus moulding" is often a false choice. The strongest approach for most hardware businesses uses both, in sequence:

1. Prototype by printing. Iterate freely until the design is right.
2. Launch by printing. Ship the first tens or hundreds of units. Real customers find problems that testing doesn't, and you can fix them for free.
3. Prove the demand. Watch reorder rates and whether the design has stabilised.
4. Tool up when the numbers justify it: and only then.
5. Keep printing for spares and variants. Long after the tool is running, printing remains the right way to make service parts, low-volume colour variants and one-off replacements that would need a separate tool.

This sequence removes almost all the risk. You never commit capital to a tool for a design that might change or a demand that might not appear.

A worked comparison

An enclosure, roughly 110 cm³, in PETG.

Printing: about ₹861 for one. At 100 units the 18% volume discount and amortised setup bring it to roughly ₹590 each: about ₹59,000 for a hundred.

Moulding: call the tool ₹4,00,000 for a part of this size and complexity, with a per-part cost around ₹90.

• 100 units: printing ₹59,000, moulding ₹4,09,000. Printing wins overwhelmingly.
• 1,000 units: printing ~₹5,90,000, moulding ₹4,90,000. Moulding pulls ahead.
• 5,000 units: printing ~₹29,50,000, moulding ₹8,50,000. Not close.

Crossover is around 800 units here. But note that the moulding column assumes the design is final and the tool is right first time. Add one tool modification and the crossover moves out by several hundred units.

When printing stays right at volume

A few situations where printing remains correct even at high quantity:

• Part geometry that can't be moulded without splitting it into pieces you then assemble: internal channels, complex lattices, deep undercuts.
• Every unit is different. Customised or serialised parts.
• Demand is lumpy and unpredictable. Spares inventories, where you might need three units this quarter and forty the next.
• The design is still evolving. A product on a rapid update cycle may never justify a tool.
• Very large parts where the tool would be enormous and the volume moderate.

How to actually decide

Get two real numbers rather than reasoning from rules of thumb:

1. A printing quote at your realistic annual volume: use the estimator, set the quantity honestly.
2. A tooling quote from a moulder, including the tool, per-part cost, lead time and the cost of a modification.

Then multiply out at your actual expected volume for the next twelve months, not your optimistic five-year projection. Compare totals. Then apply the three questions above: is the design final, is the demand proven, can you wait?

If printing is within roughly 30% of moulding on total cost, print. The flexibility is worth more than the difference, because it keeps your options open at a stage when your information is still incomplete.

Where we fit

We print. We don't sell tooling, so this comparison isn't a sales pitch in either direction.

If your volumes have grown past the point where printing makes sense, we'll say so, and you should go and get a tool quote. That's a good problem to have, and pretending otherwise would cost you money.

What we're useful for is everything before that point, and the long tail afterwards: prototypes, launch volumes, bridge production while the tool is cut, spares, variants and the parts that are too complex or too low-volume to ever justify steel.