CNC Machining Hourly Rate: Why Budgets Crash

• 1
  What a CNC Machining Hourly Rate Actually Recovers
• 2
  Five Drivers That Move the Hourly Rate
• 3
  Why the Real Invoice Exceeds the Quoted Rate
• 4
  How to Check a Rate Without Seeing the Shop's Books
• 5
  Hourly Rate by Machine Class and Job Type

A CNC machining hourly rate is a shop's way of spreading fixed and variable cost across spindle time. It is not a wage. The rate has to recover machine depreciation, floor space, power, tooling, coolant, programming, setup labor, inspection, and the risk of scrapping a part late in the run. When a buyer compares two quotes, they are usually comparing two different recovery models, not two prices for the same thing.

The useful unit is not the hour. It is the hour actually spent cutting. A 3-axis mill that runs a 20-minute cycle has a very different cost profile from a 5-axis center that finishes the same feature set in 9 minutes with one setup. The faster machine carries a higher hourly rate and often produces the cheaper part. That inversion is where most budget surprises start.

Shops quote in three common formats. Hourly rate plus material is the most transparent. Cycle-time-based pricing hides the rate and sells you the part. Feature-based pricing bundles everything into one number per part. Ask which model is being used before you argue about the rate itself. A low rate on a slow process loses to a mid rate on a fast one almost every time.

Setup is the part engineers forget. A fixture that takes four hours to build does not disappear because the cycle is short. On a 10-piece order, setup can be 60 percent of the invoice. On a 5,000-piece order it rounds to nothing. Same part, same rate, two very different unit costs.

Machine class sets the floor. A 3-axis mill, a 4-axis mill, a mill-turn center, and a simultaneous 5-axis machining center all carry different depreciation, different floor space, and different operator skill requirements. The rate has to follow the machine. You cannot run a 5-axis center at a 3-axis rate and stay open.

Material removal rate sets the ceiling. Aluminum 6061 cuts fast and tools last. Inconel and Ti-6Al-4V cut slowly, generate heat, and wear carbide quickly. The same pocket that takes 12 minutes in 6061 can take 90 minutes in Inconel. The hourly rate may not change at all, but the hours do. That is why quoting from material type alone misleads people.

Tolerance and finish add passes. Holding ±0.005 mm is not the same job as holding ±0.05 mm. Tight tolerance usually means a roughing pass, a semi-finish pass, a stress-relief pause, and a finish pass, plus more inspection. A surface finish of Ra 0.2–0.8 μm may require a separate finishing operation after the main cut. Each pass consumes spindle time at the same rate.

Geometry decides how many setups. A part that can be reached from one direction at 5 axes avoids refixturing. A part that needs four sides machined needs four setups, or a tombstone fixture, or a mill-turn platform. Every extra setup adds load, unload, re-datum, and typically a first-article check. Setup hours are billed at the same rate as cutting hours.

Volume divides the fixed cost. Programming, fixturing, and first-article inspection are one-time costs. Spread over one prototype they dominate the price. Spread over 10,000 pieces they nearly vanish. This is why a prototype quote and a production quote for the same drawing can differ by a factor of four without anyone being dishonest.

Most overruns do not come from the rate. They come from the assumptions underneath it. A quote assumes a certain number of setups, a certain cycle time, and a certain scrap allowance. When a drawing changes after the first article, or a feature turns out to be unreachable, the assumptions break and the hours grow. The rate stays the same. The invoice does not.

Thin walls and deep pockets are the classic example. A wall at 0.8 mm will deflect under normal cutting force, so the programmer has to reduce depth of cut, add support, and accept a longer cycle. None of that shows up on the drawing. It shows up in the cycle time, and therefore in the price.

Hardness variation in raw stock is another quiet driver. Two bars of the same grade can machine differently if heat treat or supplier changes. The shop may need to slow the feed to protect the tool, or add an intermediate anneal. Again, the rate does not move. The hours do.

Rework is the last one. A part that fails final inspection at 95 percent completion costs nearly as much as a fresh part, and it produces no revenue. Shops that inspect in process catch this early. Shops that inspect only at the end pay for it in scrap, and that cost is baked into every future quote they issue.

The practical fix is a DFM review before the quote is final. When manufacturability is checked first, the assumptions in the quote are grounded in a real toolpath, not a sketch. That is where most of the variance goes away.

You will rarely get a full cost breakdown, and you do not need one. You need enough information to test whether the number is coherent. Start with the machine class assigned to each operation. If a shop calls out 5-axis work at a 3-axis rate, either the rate is subsidized or the part is simpler than the drawing suggests.

Ask for cycle time per operation and setup time separately. A shop that can name both numbers has actually planned the job. A shop that only gives a lump sum has guessed. This single question separates the two groups faster than any audit.

Compare quoted tolerance capability against the drawing. If the drawing calls for ±0.01 mm and the shop quotes a standard rate, confirm they have the inspection equipment to prove it. Capability without measurement is a claim, not a process. A report on request is a reasonable ask.

Check the material and finish line items. If material is quoted separately, you can verify it against published mill pricing. If tooling or post-processing is bundled, ask what is included. Anodizing, plating, and heat treat are separate operations with separate lead times. They should not be invisible.

Finally, look at how the shop handles changes. A clear change-order process with re-quoted hours is a sign the original quote was real. A shop that absorbs every change silently is either very efficient or building the cost into the next quote. You will not know which until the second order.

One more check: ask what happens if the first article fails. A shop with in-process monitoring will catch it early and tell you. A shop that only inspects at the end will tell you late. Both charge for the time. Only one gives you a chance to fix the drawing.