Injection Molding Cost Drivers: Tooling, Volume, Cycle Time

Every injection-molded part carries two costs that behave in opposite ways: a one-time tooling investment and a per-part piece price. The piece price is built from material, which is resin price times shot weight; from machine time, which is machine cost times cycle time divided by cavity count; and from secondary operations such as trimming, assembly, or painting. The tooling is amortized, meaning the mold cost is divided across the total quantity of parts the tool will produce. That division makes quantity the single largest lever on unit price: at low volumes the tool dominates the cost of every part, and past the low tens of thousands of parts the tooling share fades until material and cycle time set the floor. This page is the cost companion to the injection molding process guide. It explains each driver in relative terms, how it scales and which direction it pushes, because actual prices depend on the part, the tooling class, the supplier, and the market at the time.

The drivers interact, and understanding the interaction is what lets an engineer read a price or shape a part that costs less. Wall thickness moves cooling time, which grows with the square of thickness. Undercuts add mechanisms to the mold and cap how many cavities it can hold. Cavity count divides machine time but multiplies tool cost. Resin class sets the material tier, and tool class sets how many shots the tool can deliver before it wears out. The sections below take the drivers one at a time, then rank the design levers that move them, and close with what a supplier actually prices from when quoting a molded part.

Two cost structures in every molded part

Cost guides from molders and design-for-manufacturing sources agree on the skeleton. Part cost equals material cost, plus processing cost divided by the number of parts made per cycle, plus tooling cost divided by the production volume. The three terms are independent of each other: material is bought by weight, processing is bought by machine hour, and tooling is bought once. A supplier reading a part prices all three, and a designer reading a drawing moves all three, which is why the same part can cost meaningfully different amounts from the same press depending on how it was drawn.

Total program cost then follows a simple shape: tooling as a capital item, plus the piece price multiplied by the volume the program will actually run. Once the mold is paid for, the per-part cost is dominated by machine time and raw material. Published estimates from molding cost guides put the shares at production volume near these bands: material roughly 20 to 50 percent of per-part cost, machine and processing roughly 20 to 50 percent, secondary operations roughly 5 to 20 percent, and logistics 1 to 10 percent. The bands are wide because they shift with volume and part type, but their existence is the point: no single term wins everywhere, and the lowest-cost decision on one term can raise another.

How tooling amortization works

Tooling cost per part equals the total mold cost divided by the production volume, and that one division explains most of how injection molding prices behave. The mold cost is fixed the day the tool is cut. The volume is not. Everything that follows in this section is the consequence of spreading a fixed cost over a moving quantity, and it is the mechanism behind minimum order quantities, behind the steep per-part price fall that molding is known for, and behind the plateau where ordering more stops helping.

Published volume-tier estimates from molding cost guides show the tooling share of per-part cost at roughly four-fifths around 1,000 units, falling to a minor share by around 50,000 units, with the material share rising as the tooling share falls. The curve is steep and then flat. One molder describes it as a knee point where the process turns highly cost-competitive, and the knee exists because that is where the amortization term crosses from dominant to minor.

For example, hold the mold cost fixed and move only the quantity. At 10,000 parts, each part carries one ten-thousandth of the tool; at 100,000 parts, each carries one hundred-thousandth, a tenth as much. Published comparisons reflect exactly this arithmetic: at a few hundred pieces, the per-part price is commonly cited at four to six times the price the same mold delivers at around 5,000 pieces, and doubling an order from 50,000 to 100,000 parts saves only a few percent, because by then the tooling term has faded and material and machine time set the floor. Quantity earns its savings early and then stops earning them, which is a fact worth having before any volume negotiation.

Which tool you divide by the volume matters as much as the volume itself. Aluminum tools cost markedly less than steel, with published claims clustering around half to two-thirds of an equivalent steel build, but they live a fraction as long, so aluminum wins the per-part tooling arithmetic at low volume and loses it at high volume. Hardened steel is the only class for production in the millions of shots. The tool classes, their shot-life bands, and the SPI Class 101 through 105 convention that summarizes them are covered on the rapid tooling page; here the class choice is one input to the amortization arithmetic. The same fixed-tooling-versus-per-part trade governs metal parts, and it is compared head to head on the die casting vs injection molding page.

Why minimum order quantities exist

A minimum order quantity is the volume at which the supplier’s pricing arithmetic works, and it is built from three recurring costs: tooling recovery, machine setup, and material minimums. As one molder puts it, the number is not pulled out of thin air. Below it, the fixed costs of the run are never recovered across the parts in it.

The recurring setup deserves a close look, because it repeats on every run regardless of order size. The mold must be retrieved, cleaned, installed, and aligned; cooling and hydraulic lines must be connected; hygroscopic resin must be dried before the first shot; the previous material must be purged from the barrel, which discards material; and process parameters must be dialed in until the first article passes. Published setup durations cluster at roughly two to eight hours of skilled work, during which the press produces no revenue-generating parts. A run too short to absorb that setup is unprofitable at almost any piece price, which is the root cause of minimum-order behavior rather than any sales policy.

Material minimums add order granularity from the other direction. Resin is sold in bag and box lots, commonly 25 to 50 kg bags, and specialty resins can carry larger minimums. A small part cannot be molded from a fraction of a bag, because a molder cannot buy half a bag. The purchasable lot, not the part weight, sets the smallest sensible order.

As commonly published direction rather than a norm, aluminum and prototype tooling tends to pair with minimums in the hundreds to low thousands, single-cavity steel with a few thousand, and multi-cavity production tooling with the tens of thousands. Absolute floors vary widely by supplier and region, so treat the bands as shape, not as commitments. Buyers do have real levers: pay tooling separately from the piece price, which removes the amortization term from each part; use a family mold that runs several similar parts of the same resin in one tool, so the minimum applies to total output; specify standard mold bases, published at roughly 15 percent below fully custom bases; offer flexible scheduling or a stock-and-release agreement so the molder can fill press gaps; and choose aluminum tooling for runs under about 5,000 parts. None of these change the arithmetic. They change which bucket the fixed costs land in.

Where molding overtakes tooling-free routes commonly sits in the hundreds to low thousands of parts, and higher when side actions or tight tolerances are involved. The crossover model itself is worked through on the injection molding vs 3D printing page, and the broader tooling-versus-per-part trade across processes on the low-volume manufacturing page.

Cycle time: the dominant per-part driver

Once tooling amortizes, machine time is the term a designer can still move, and the molding cycle belongs to cooling. Published estimates put cooling at typically half to three-quarters of the total cycle, and more for thick-walled parts. Filling and packing take a few seconds, and mold open, ejection, and reset take a few more. Any effort to shorten a cycle that does not touch cooling is rearranging the small share.

The physics that matters is that cooling time grows with the square of wall thickness. Double the wall and the cooling time roughly quadruples; triple it and the cooling time rises roughly ninefold. The thickest local section dictates the cycle for the entire part, so one thick boss can hold an otherwise thin housing in the press for extra seconds on every shot. The consequences are published in design-for-manufacturing analyses: a reduction of about 1 mm in average wall thickness cuts cycle time on the order of 20 to 40 percent, and halving a 4 mm wall to 2 mm cuts cooling by roughly three-quarters under the squared law.

Machine cost per part equals the machine’s hourly cost times the cycle time, divided by the cavity count, so a 60-second cycle costs twice a 30-second cycle in machine overhead on the same press. Press size follows the part’s projected area and the pressure needed to fill it, so a bigger part forces a bigger press, and larger presses carry higher machine-time cost per hour. Cooling design is a cycle lever independent of the part itself: clean, well-placed cooling channels are published to cut cycle time by 10 to 20 percent, and conformal cooling channels are claimed at 20 to 40 percent on suitable geometry. What cannot be done is pushing the cycle faster than the part can solidify, because under-cooling buys warpage, sink, and dimensional variation. Cycle reductions come from design and cooling, not from hurrying the plastic.

The wall and rib rules that keep sections thin and uniform are carried as numbers on the design rules page. Here the point is what those rules are worth: because cooling is the majority of the cycle and follows thickness squared, wall thickness is the single biggest piece-price lever available to a designer, and it is free before the tool is cut.

Material: tiers, shot weight, and waste

Material cost per part equals the resin price per unit weight times the shot weight, adjusted for scrap, and the shot weight is not the part weight. It is the part plus the runner plus startup and shutdown scrap. In a cold-runner mold, published measurements put 15 to 40 percent of every shot into the runner, which is reground or discarded. A hot runner keeps that material molten and cuts the waste to under 2 percent, at the price of a heated manifold, nozzle tips, and temperature-control zones added to the tool.

Resin prices fall into tiers whose ordering is stable even though the absolute prices move with the petrochemical cycle. Commodity plastics such as PE, PP, and PS are the lowest tier. Engineering resins such as ABS, polycarbonate, nylon, and acetal run a mid tier, typically a small multiple of commodity pricing. High-performance resins such as PEEK, PEI, and LCP sit an order of magnitude or more above commodity. The resin-by-resin selection detail lives on the injection molding materials page.

Additives add percentages to the resin price: color at roughly 5 to 15 percent, flame-retardant and UV-stabilized packages at roughly 10 to 30 percent, and glass or mineral reinforcement at roughly 20 to 50 percent above the base resin, per published design-for-manufacturing cost guides. Filled resins also raise tool cost, because glass and mineral fibers are abrasive: they wear cavities and gates faster, push tooling toward hardened steel, and shorten maintenance intervals. A resin choice is never only a material-line decision; it is a tooling decision too.

Where the resin and the application allow it, regrind is the standard offset: published guidance puts tolerable regrind fractions near 15 to 30 percent for commodity resins and 10 to 20 percent for engineering resins, with restrictions where virgin documentation, optical clarity, or color consistency is required. Hot runners repay their added tool cost above roughly 50,000 total shots in commonly published break-even guidance, earliest with expensive resin, high cavitation, and long runs, and offset by added maintenance exposure in heaters, thermocouples, and valve gates. The quiet trap is over-specification: a lower resin class substituted where the duty genuinely allows can cut material cost by double-digit percentages, but the substitution has to survive the real loads, temperatures, and chemical exposure, and the honest way to find that out is a material review, not a hope.

Cavity count: the multiplier that cuts both ways

Cavities divide machine time and multiply tool cost, and the multiplication is sub-linear. A 4-cavity mold typically costs about 2.5 to 3 times a single-cavity mold, an 8-cavity mold about 4 to 5 times, and doubling from one cavity to two adds only about 50 to 70 percent to the tool price, per several independent cavity-count guides. The division, by contrast, is exact: four cavities produce four parts per cycle, so the machine time carried by each part falls to about a quarter.

For example, compare a single-cavity tool and a 4-cavity tool at 100,000 parts, using the published multipliers. The 4-cavity mold costs about three times as much, so the tooling premium is two additional tool costs spread across 100,000 parts, a vanishingly small addition per part, while machine time per part falls to about a quarter. Published comparisons put the per-part molding cost of a 4-cavity tool more than 60 percent below the single-cavity equivalent once volume clears the tooling premium, and a part that permits a very high cavity count can run an order of magnitude lower in machine time per piece than the same part molded one-up.

The returns diminish and then reverse. Sixteen cavities may save only marginally more per part than eight. More cavities demand a larger shot, a larger mold, and a larger press, and part size caps the count outright, because the tool must fit within the tie-bar spacing and platen dimensions of an available machine. Every added cavity is another cavity that can stop the whole tool: at a 1 percent per-cavity problem rate, a 16-cavity mold carries roughly a 15 percent chance that some cavity stops it during a run. Cavities must also fill in balance, or scrap rises with the count. And undercuts cap cavitation before anything else does: a part that needs side pulls on all four sides is effectively a one-cavity part, while undercut-free caps and closures run production molds with more than 128 cavities. The right cavity count is an economics calculation against annual volume, not as many cavities as will fit.

Mold complexity and class as cost multipliers

What makes a mold expensive is concrete: the mold base; the cavity and core work itself, milling, EDM, polishing, and texturing; design and engineering effort including flow analysis and trial runs; the runner system; and any side pulls, lifters, or unscrewing mechanisms. Complexity raises cost through each of these paths, because more surfaces, more undercuts, and more specialty features each add machining and assembly time.

Side actions are the clearest example. Each slider or lifter is a step increase rather than a gradient: published claims put a single slider at roughly 15 to 30 percent added mold cost, and slider-equipped molds at roughly 40 to 60 percent above an equivalent two-plate tool. Steel grade moves cost on a relative index with pre-hardened P20 as the 1.0 baseline: published tooling guides put hardened H13 near 1.5 to 2.0 and stainless S136 near 2.0 to 2.8. The steel itself is a minor share of the mold cost; what the grade buys is heat treatment, harder machining, and polishing labor, so the grade cascades through the build rather than appearing as a line item. Aluminum sits below all of these and is quicker to cut and to modify, which is why it anchors the low-volume end of the tooling ladder described on the rapid tooling page.

Finish and tolerance behave the same way. A mirror finish in the SPI A classes takes hours of skilled hand polishing and is published to add on the order of 30 to 50 percent to cavity cost against a textured or machined finish; downgrading a non-cosmetic surface to a functional finish removes that labor. Tight tolerances raise cost through the steel as much as through the process: a part feature held to ±0.001 in requires cavity steel held to roughly ±0.0005 in, plus tighter process control and more inspection. Blanket tight tolerances multiply that cost across every dimension on the drawing, so the standard remedy is to localize precision to mating, sealing, and locating features and leave the rest at general tolerance; one molder cites tolerance-grade selection for non-mating surfaces in the ISO 20457 series as worth roughly 10 to 15 percent in processing and scrap. The tolerance classes themselves are detailed on the tolerances page.

A hot-runner system adds a discrete block of cost, the manifold, nozzles, and control zones, while a cold-runner mold is published at roughly 20 to 40 percent less to build; which one wins is the arithmetic between runner waste, resin price, and total shots. Mold class, from prototype Class 105 through production Class 101, is the industry-convention shorthand that bundles steel grade, guidance systems, and expected maintenance into a life-expectation band, and the class-by-class breakdown is laid out on the rapid tooling page.

The design levers, ranked by effect

Nearly every lever below is free before steel is cut and expensive after, which is the entire argument for design review ahead of tooling. Ranked roughly by published effect on unit price, the levers are:

• Thin, uniform walls, because cooling follows the square of the thickest section.
• Eliminating undercuts and side actions where function allows, because each is a tool-cost step and a cap on cavitation.
• Setting the cavity count against the annual volume, not against the first purchase order.
• Localizing tight tolerances to the features that need them.
• Right-sizing the finish, functional where the surface is not seen.
• Right-sizing the resin class to the actual duty.
• Right-sizing the tool class to the expected volume.

Two of these deserve expansion. Ribs instead of thick walls are the classic double saving: a rib stiffens without the sink and cycle cost of a thick wall, so the part uses less material and cools faster, and coring out thick sections does the same from the inside. The cycle follows the thickest local section, so a part with one solid boss pays for that boss on every shot. On color and finish, natural or standard-black resin costs less than custom color matching at low volume, a textured surface can substitute for gloss where cosmetics allow and hides minor sink read-through, and buying resin in pallet quantities is published to save 10 to 20 percent at the cost of carrying inventory.

Design review before steel is cut has the largest impact on total program cost of anything on this list, a point molders make consistently. The specific savings ratios in bureau marketing are single-source claims, so treat the direction as solid and the numbers as case studies rather than expectations. The mechanism is not in doubt: a change caught in review costs a drawing revision, and the same change after steel is cut costs welds, remachining, and re-texturing, or a new insert. Engineering change orders on cut steel are the classic hidden cost, paid twice, once for the original error and once for the fix. Family molds, which amortize one tool across a set of similar same-resin parts, extend the same logic from one part to a whole product.

What a supplier prices from, and why quotes differ

A supplier prices a molded part from a defined set of inputs, and every gap in those inputs becomes an assumption, which becomes a price difference. The geometry arrives as a 3D model, typically STEP, with units stated explicitly, because a unit error rescales the part by a factor of 25.4 and quietly invalidates the estimate. The drawing, not the model, carries intent: critical dimensions, tolerance classes, finish callouts, and which features are truly critical rather than left at general tolerance.

The rest of the package: the resin grade and color, and whether regrind is acceptable; the expected annual volume and the total program volume, because together they drive cavity count and tool class; cosmetic surfaces called out by finish class; secondary operations declared up front so they sit inside the price rather than after it; and undercuts and side-action features flagged explicitly, because no other geometry callout moves the tooling estimate as much.

Two suppliers quoting the same part can legitimately produce different numbers, because they assume different tools: a different mold class, a different cavity count, a different tolerance interpretation, a different finish, a different resin policy, different exclusions. Some prices assume bare molded parts and leave trimming, assembly, packaging, or freight outside, a gap molders themselves warn about. The honest comparison is of assumptions, not totals. The process-generic version of that anatomy is laid out on the how quotes are calculated page.

Managing cost across a program

A few program-level habits work with the arithmetic rather than against it. Batch reorders into fewer, larger runs, so each setup amortizes across more parts. Offer flexible delivery windows, which let a molder fill press gaps and are a legitimate price lever rather than a favor. Use stock-and-release agreements to hold finished inventory against a schedule. Cover a demand ramp with bridge tooling while a production tool is built, a strategy worked through on the tooling page. Watch the quiet post-tooling costs that published budget guides warn about: engineering change orders on cut steel, mold maintenance at a small percentage of mold cost annually, source-dependent, mold storage, first-article inspection, and compliance testing, which together can inflate a budget by a published 20 to 40 percent if unmanaged. Logistics is a real but minor share, 1 to 10 percent of landed cost, and landed cost, not the ex-works number, is the honest basis for comparing suppliers, especially across borders where duties and freight modes differ.

The one-line summary: injection molding cost is a fixed tool divided by a quantity, plus a per-part price set by resin, cooling, and cavities, plus whatever happens after the press. Quantity rebalances the first term, design moves the second, and discipline in scoping controls the third. The language of molding economics is collected in the manufacturing glossary.