Injection Molding vs CNC Machining Cost Per Part

The short answer: CNC machining has no tooling cost and a per-part price commonly quoted between $12 and $500 depending on complexity. Injection molding charges $2,000 to $150,000 or more for the mold first, then $0.50 to $5.00 per part. The crossover is not a fixed number, and the widely repeated "about 1,000 units" rule is wrong by a factor of roughly 40. Run the arithmetic yourself: break-even units equals tooling cost divided by the per-part saving. On real quotes that lands anywhere from 71 units to 11,429.

That spread is the whole reason this decision keeps getting made badly. Two teams read the same rule of thumb, one commits $60,000 to steel for a product that will sell 4,000 units, and the other machines 9,000 parts one at a time because someone said molding needs high volume. Both are following the same advice. The rule is not wrong so much as meaningless without your two numbers in it.

The two costs you are actually comparing

CNC machining cuts each part from a solid block. There is no tool to pay for, so the first part and the thousandth cost roughly the same, minus a modest setup amortization. Published ranges for machined plastic parts sit around $50 to $500 per part for anything with real geometry, and lower, in the $12 to $30 range, for small simple shapes at moderate quantity. Lead time is days.

Injection molding front-loads everything into the mold. Once that steel exists, each shot produces a finished part in seconds and per-part cost for a typical consumer part lands around $0.50 to $5.00, falling toward pennies in a multi-cavity production tool. The catch is what the tool costs and how long it takes to cut. Our guide to plastic injection molding companies and custom plastic molding services covers how tooling quotes are built and why the same drawing comes back with a 20x spread.

So the comparison is not process against process. It is a fixed cost against a variable one, and the only question that matters is at what quantity the fixed cost stops being the expensive choice.

The break-even formula, and what it actually returns

Break-even units equals tooling cost divided by (CNC price per part minus molded price per part). Nothing more sophisticated is needed, and both inputs come straight off your quotes.

Here is the same formula run against five real tooling tiers under two scenarios. Scenario A is a part with pockets and features that make it slow to machine: $45 per part on CNC, $2.50 molded, so the saving is $42.50 a part. Scenario B is a small simple shape that machines fast: $12 per part on CNC, $1.50 molded, a saving of $10.50.

Read the two right-hand columns against each other. The break-even for the same decision ranges from 71 units to 11,429 units, a spread of about 160x, purely from which tool you were quoted and how expensive your part is to machine. That is why a single industry-wide crossover number cannot exist, and why quoting one to your finance team is a good way to get the decision wrong in either direction.

Two corrections worth making before you trust your own result. The molded price you plug in must be the price at the quantity you are actually ordering, not the price at 100,000 units, because those are different numbers. And the CNC price should be the quantity-adjusted one, since machining does get modestly cheaper per part once setup is amortized across a batch.

Is CNC machining cheaper than injection molding?

Below break-even, yes, and usually by a lot. For under a few hundred units of most parts, machining wins on total cost and wins even harder on time, because there is no four to twelve week tool build in front of you. Above break-even, molding wins decisively and the gap widens with every unit, since you are comparing a per-part cost of a couple of dollars against one of tens of dollars.

The point people miss is that this is a total-cost comparison over a stated horizon, not a per-unit one. Machining looks expensive per part and cheap per project at low volume. Molding looks cheap per part and expensive per project until the run is large enough. Pick the horizon first, three years of realistic demand is a reasonable default, then compute.

What about the option in the middle?

There is a third route that gets skipped, and it changes the math more than most people expect. Bridge tooling is a cheap aluminum mold, often $2,000 to $10,000, built to produce a few thousand parts rather than a few hundred thousand. Protolabs, one of the few vendors publishing a number at all, lists mold cost starting at $1,495 and guarantees its aluminum prototyping tools for at least 2,000 shots, with standard lead times as fast as seven days.

That option collapses the risk in the decision. Instead of choosing between machining 5,000 parts at $45 each and betting $60,000 on steel, you spend $6,000 on aluminum, get molded parts in weeks, sell through the first few thousand units, and cut production steel once demand is real rather than forecast. Note also that molded and machined parts are not identical objects: molding needs draft angles, uniform wall thickness, and ribs rather than solid sections, so a design that was machined usually needs a genuine revision before it can be molded. Doing that revision at the aluminum stage is far cheaper than discovering it after the steel is cut.

How many units do you need to justify injection molding?

Compute it rather than looking it up. Take the tooling quote, divide it by the difference between your CNC per-part price and your molded per-part price, and compare the answer to demand you would actually commit to buying. On the numbers above that threshold sits somewhere between 71 and 11,429 units. If you have no quotes yet and need a placeholder for a first conversation, a few hundred units for a simple part on aluminum tooling and a few thousand for a complex part on steel is closer to reality than a single blanket figure.

One caveat that catches teams out: crossing the break-even on paper is not the same as having the cash. Tooling is paid up front, typically as a deposit before the tool room starts and the balance at sample approval, while the per-part saving arrives slowly across a production run. A decision that is correct over three years can still be wrong this quarter.

Which process should you use at each stage?

Ranges are planning guides drawn from what US and overseas shops commonly quote in 2026, not promises. Your own two quotes decide the answer.

What should you put in the RFQ so the quotes are comparable?

Most of the confusion in this comparison comes from quotes that describe different things. Send the same brief to every shop: the STEP file, the resin with any filler percentage, the expected annual volume and the number of years the part stays in the catalog, the SPI mold class you want quoted, the cavitation, the color or masterbatch requirement, and the tool ownership terms. Ask for the molded price at two or three quantities, because the second number is where you learn the shape of the curve.

Then send that brief to eight or ten shops rather than three. Because almost nobody publishes injection molding prices, the only way to see the real spread is to collect it, and the spread on the same part across honest molders is routinely 10x. That is exactly the tedious part worth automating, which is what structured RFQ software is for once you are past a handful of vendors. If you are early and the constraint is order size rather than price, low MOQ manufacturers is the better place to start.

What does the decision look like written down?

Whichever way the math falls, write it down. A one-page cost model with the crossover volume, the two per-part prices, the tooling quote, and the demand assumption behind all of it is what stops the decision getting re-litigated every quarter, and it is the document you will want when someone asks you to justify the spend six months from now. If it has to go in front of a committee, turning that spreadsheet into a deck automatically is quicker than rebuilding the numbers by hand.

Then keep the assumption visible. The break-even you computed is only valid for the demand figure you fed it. When demand changes, the answer changes, and the whole point of writing the formula down rather than a conclusion is that you can re-run it in thirty seconds.

How do you find molders and machine shops that will actually quote you?

The practical bottleneck is not the arithmetic, it is getting comparable numbers from shops that are genuinely set up for your part. Presses are sized to part volume, tool rooms specialize by steel and part class, and a machine shop that is excellent at aluminum housings may not run engineering plastics well. Describe the part, the resin, and the volume in plain English and the AI sourcing agent shortlists CNC machining companies and injection molders that match, with vetting evidence, capabilities, lead times, and quotes side by side. Broader options across processes sit in our contract manufacturers guide.