CNC Processing Price: What Actually Drives the Number

• 1
  Key takeaways
• 2
  Where two quotes for the same part diverge
• 3
  Material choice moves the CNC processing price first
• 4
  Tolerance and surface finish drive cycle time
• 5
  Setup, fixturing and part count
• 6
  Machine capability and shop rate
• 7
  Volume, lead time and the cost of speed
• 8
  Finishing, inspection and documentation
• 9
  Seven things to check before you compare quotes
• 10
  Step by step: getting a comparable CNC processing price

Material is the first line item in any CNC processing price, and it works on two levels. The stock itself has a price per kilogram, and the grade decides how long the cutter stays in the cut. Aluminium 6061 cuts fast, throws chips cleanly and tolerates aggressive feeds. Stainless 316 work-hardens under a dull tool, so feed and speed have to be managed or the surface tears.

Titanium TC4 (Ti-6Al-4V) and Inconel sit at the far end. They hold strength at temperature, which is exactly why they are slow to cut. Tool life drops, coolant demands rise, and cycle time can be three to five times that of 6061 for the same pocket. If the drawing does not need that heat resistance, the material is costing you money for nothing.

Stock form matters too. Plate is usually cheaper than bar for flat parts because there is less to remove. Near-net forgings or castings reduce roughing time but add tooling cost and lead time, and they only pay off above a few hundred parts. For one-off prototypes, bar or plate is nearly always the cheaper route.

A practical check: list the two or three properties the part truly needs, then pick the cheapest grade that meets them. Hardness, corrosion resistance and weight are the usual three. Everything else is a preference, and preferences are what push a quote up.

Tolerance is where engineers most often over-specify. A general block tolerance of ±0.1 mm is easy. A ±0.005 mm bore is not just a tighter machine setting. It means a finish pass with a smaller stepover, a temperature-stable setup, and a CMM check that has to be scheduled.

The same logic applies to surface finish. Ra 3.2 μm comes off the machine with a normal pass. Ra 0.8–1.6 μm needs a controlled finishing pass. Ra 0.2–0.8 μm usually needs a second operation, sometimes lapping or polishing by hand. Each step adds time per part, and hand work does not scale with volume.

A useful rule: apply tight tolerance only to the features that mate with something else. Bearing bores, seal faces, dowel holes and spigots earn their tolerance. A clearance hole for a cable tie does not. Marking the critical features on the drawing, or listing them in the RFQ, lets the shop spend time where it matters.

If a feature can be opened from ±0.005 mm to ±0.02 mm without hurting function, say so. On a typical aluminium housing, that single change can remove a finishing pass and a CMM check, and the saving shows up directly in the CNC processing price.

Every part needs a setup before the first chip. The machine has to be cleared, the fixture mounted, tools loaded and offsets proven. On a 3-axis machine, a part with features on five faces may need three or four setups. Each one adds an hour or more of skilled time and a fresh chance of position error.

A 5-axis machine can often reach the same part in one setup. The hourly rate is higher, but the setup hours drop and the position error between faces disappears. For a one-off complex bracket, 5-axis is frequently cheaper overall. For a simple plate with holes, 3-axis wins.

Fixturing is the quiet cost. Soft jaws and standard vises are quick and cheap. A dedicated fixture for an odd casting takes design time and a fixture build, which is why the first article of a new design always costs more. Once the fixture exists, the per-part cost falls.

Volume changes everything. Spreading 4 hours of setup over 5 parts adds 48 minutes per part. Spread the same 4 hours over 500 parts and it adds 29 seconds. This is why a shop cannot quote a per-part price without knowing the quantity band.

Shop rate is not a single number. A 3-axis mill and a simultaneous 5-axis machining center carry different hourly costs, because the machine, the tooling and the operator skill all differ. A quote that hides the machine mix is hard to compare with another quote.

The question to ask is not which machine is cheaper per hour. It is which machine finishes the part in fewer total hours. A part with undercuts, deep pockets or angled ports may need two 3-axis setups plus a fixture that a 5-axis machine avoids. The higher rate can still produce the lower total.

Capacity matters as well. A shop running 127 high-precision machines, including 16 simultaneous 5-axis centers and 16 mill-turn centers, can route a job to the machine that suits it rather than forcing it onto whatever is free. That routing decision is where a good shop earns its margin.

For large parts, size limits decide the quote. A maximum processing size of 4,000 mm rules out many suppliers completely. Check the travel of the machine before you assume a shop can take the job, especially on long extrusions or frame rails.

Unit price falls with volume, but not in a straight line. The first break comes when setup is amortised. The second comes when the shop can batch parts and run lights-out. The third comes when material is bought in a larger quantity at a better rate.

Short lead times cost money. To ship in 3–5 days, a shop has to interrupt its schedule, run a setup that was planned for later, and sometimes use express material delivery. That is real cost, not a penalty invented to pad the invoice. If the date is flexible, say so, and the number usually improves.

A common mistake is to ask for a 10,000-part price and then order 50. The quote assumes batching, dedicated fixtures and bulk material. At 50 parts those assumptions break, and the shop either loses money or the price is renegotiated. Give the real annual demand and the real first order.

It also helps to separate prototype, pilot and production quantities in one RFQ. A shop can then quote each band honestly and tell you where the price steps down. That is more useful than one blended number that fits none of your actual orders.

Finishing is often quoted separately because it involves different equipment and sometimes an outside vendor. Anodizing, plating, powder coating and black oxide all add handling and turnaround time. Bead blasting, tumbling, brushing and polishing are cheaper but still add a step per part.

Laser marking is a small line item with a real constraint. Minimum character height is 1.5 mm. If your part number or traceability code needs smaller text, the marking step has to change, and that change belongs in the original quote discussion rather than in a later argument.

Inspection depth is another line. Sample checks cost little. Full inspection of critical dimensions adds CMM time, and 100% inspection before shipment adds handling. Raw material checks, in-process monitoring and final inspection reports can all be requested, and each one has a cost.

Certification requirements also shape the price. Work under ISO 9001:2015, IATF 16949:2016, ISO 13485:2016 or ISO 27001:2022 carries documentation overhead. That overhead is part of a medical or automotive quote, and it is why those quotes rarely match a general machining quote line for line.

Quotes are only comparable when they describe the same scope. Two numbers on a screen tell you nothing if one includes anodizing and the other does not, or if one assumes ±0.1 mm and the other ±0.005 mm. Build a small comparison sheet and fill it in for every supplier.

Start with the drawing revision. Then note the material grade and stock form, the tolerance on critical features, the finish, the inspection level, the quantity band, and the delivery date. If any cell is blank, ask before you compare totals.

It also pays to ask what happens after the first article. Is the price held for repeat orders? Does the fixture stay on the shelf? Are the inspection reports archived? These answers separate a shop that plans for production from one that quotes each job from scratch.

Finally, treat the DFM feedback as part of the quote. A shop that flags a deep slot, a thin wall or an impossible thread before cutting is saving you a scrapped batch. That feedback can change the design, and a changed design can change the CNC processing price in your favour.