Material Cost in Machining: The Share Most Buyers Underestimate
Ask a buyer what dominates the price of a machined aluminum part and most will say machine time. Ask a shop that has just paid for a 40 kg billet to produce a 3 kg housing and the answer changes. Material is 25-55% of a machined part price depending on geometry, and on hogged-out aerospace-style parts it can pass 60%.
The reason is not the price of aluminum. It is how much of the plate you buy ends up as chips. That ratio has a name, and once you can estimate it, you can predict a quote before it arrives.
The share depends on geometry, not on the metal index
Two parts can weigh the same and carry very different material bills. A 0.9 kg manifold block cut from 120 × 90 × 40 mm plate starts as 2.6 kg of stock. A 0.9 kg turned bushing starts as 1.0 kg of bar. Same finished weight, 2.6 times the material cost on the block, because the block pays for the corners of a rectangle that the bushing never had.
That is the whole mechanism. Material cost is driven by the smallest stock envelope that contains the part, plus machining allowance, plus the saw kerf, plus whatever the mill sells in standard increments. A part that fits in 100 mm of stock may be cut from 110 mm because plate comes in 10 mm steps and the next size down leaves no cleanup allowance.
For 6061 plate in Shenzhen, expect USD 3.60-4.20/kg for common thicknesses, with 7075 at USD 7.50-9.00/kg and 304 stainless bar at USD 5.50-6.80/kg. Those prices matter less than you think. A 12% move in the aluminum index on a part with 30% material share is a 3.6% price move — noticeable, not decisive. A change in geometry that takes the ratio from 2:1 to 4:1 doubles the material line.
Buy-to-fly: the ratio that sets the bill
Buy-to-fly is stock weight divided by finished weight. Divide it into the finished weight and you have the material cost per part; multiply by the price per kilogram and you have the money.
A bracket needing 4.6 kg of plate for a 1.2 kg part has a ratio of 3.8:1. At USD 4.00/kg, material is USD 18.40. If the part sells at USD 42, material is 44% of the price. Push the ratio to 6:1 and material becomes USD 28.80, which either eats 10 points of margin or adds 25% to the price.
| Part type | Stock form | Finished weight | Stock weight | Buy-to-fly | Material share of price |
|---|---|---|---|---|---|
| Turned bushing, Ø40 × 60 mm | Round bar | 0.35 kg | 0.78 kg | 2.2:1 | 22% |
| Thin cover, 200 × 150 × 8 mm | Plate | 0.50 kg | 1.05 kg | 2.1:1 | 30% |
| Manifold block, 120 × 90 × 40 mm | Plate | 0.95 kg | 2.60 kg | 2.7:1 | 38% |
| Long rail, 600 × 40 × 30 mm | Bar or extrusion | 1.40 kg | 2.30 kg | 1.6:1 | 26% |
| Pump housing, 220 × 160 × 75 mm | Billet | 2.80 kg | 9.40 kg | 3.4:1 | 52% |
| Five-axis impeller blank, Ø120 mm | Billet | 0.60 kg | 4.20 kg | 7.0:1 | 61% |
The last row is the one that surprises buyers. A part with 7:1 buy-to-fly is not being overcharged on material; it is paying for the fact that most of the billet leaves as chips, and the chips sell for a fraction of what the billet cost.
See also: Machining Cycle Time Calculation: A Worked Example on an Aluminum Bracket and Air vs Sea Freight Cost for Machined Parts: When Air Actually Wins.
Plate, billet, or near-net: three ways to buy the same part
Stock form is a design decision as much as a purchasing one.
Plate is the default for prismatic parts up to about 150 mm thick. It is cheaper per kilogram than billet, comes in standard thicknesses, and needs minimal facing. Its limitation is that you buy the full rectangle even when the part is triangular.
Billet or block stock suits thick sections and parts that need uniform grain structure. It costs 10-20% more per kilogram and usually needs more facing, but it lets you size the envelope closely.
Near-net forms — extrusions, castings, forgings — cut buy-to-fly to 1.2-1.8:1 and can remove 30-50% of the material cost. The trade-off is tooling: an extrusion die runs USD 1,500-4,000 and takes two to four weeks, which only pays back above roughly 2,000-3,000 pieces or where the geometry is repetitive.
For a 200-piece run, plate wins almost every time. For a 20,000-piece run of the same bracket, an extrusion profile priced at USD 1,000/t over billet will beat it by 15-25% on total cost.
Offcuts, chips, and the value you do not get back
Shops recover some material value, and buyers sometimes assume that recovery should be credited. It usually is not, and here is the arithmetic.
Chips from 6061 machining sell to a recycler at roughly 35-45% of ingot price, less transport. A shop generating 3 kg of chips per part on a 200-piece run ships 600 kg of swarf, worth perhaps USD 700-900. Against a material bill of USD 3,700, that is a 20-25% recovery — enough to matter, not enough to change the quote.
Solid offcuts are different. A 200 × 300 mm remnant from a plate can be used for another job if the shop has a program that fits it. Shops that run families of parts do credit this informally, and you can ask for it on a long program: "return or credit reusable remnants over 1 kg." Do not expect a line item for swarf.
Why pocketing costs you twice
A deep pocket removes material, and buyers often read that as saving weight and therefore money. It does the opposite on both counts.
First, the pocket does not reduce the stock you buy. A 30 mm deep pocket in a 40 mm plate still requires 40 mm of plate; the removed volume was never optional stock, it was the envelope. Second, removing it consumes machine time at 120-200 cm³/min. A 100 cm³ pocket costs roughly 40-50 seconds of roughing plus finishing on the walls and floor, and it may force a smaller cutter with a lower removal rate as depth increases.
A 15 mm deep pocket cut with a 10 mm end mill reaches 1.5× diameter in depth and cuts efficiently. A 60 mm deep pocket needs a long-reach tool, which means reduced radial engagement, more passes, and often 2.5-3 times the time per cubic centimeter. Deep pockets are where material cost and machine cost both rise at once, which is why a DFM review often suggests a shallower pocket with a bolt-on cover.
If you want a buy-to-fly estimate on a specific design before you commit, send the model to contact and the quote will show the stock size and the material line separately, so you can see what the geometry is costing.
Frequently asked questions
What is a normal buy-to-fly ratio for a machined part?
Between 1.5:1 and 4:1 for most prismatic and turned parts, with 2.5:1 a reasonable average. Anything above 5:1 is worth a design review, because the material line and the roughing time are both rising faster than the finished weight justifies.
Should I ask for a credit on the chips?
No. Swarf typically recovers 35-45% of ingot price before transport, which works out to 20-25% of the material bill on a chip-heavy part. Asking for it back is not standard and rarely wins anything. Asking for reusable solid remnants to be returned is more realistic.
Is billet always more expensive than plate?
Per kilogram, yes — usually 10-20% more. Across a whole part it depends on the envelope: a closely sized billet can beat a plate that forces you to buy a larger rectangle. Compare the total stock weight, not the unit price.
Does a lighter part always cost less?
Rarely. Removing material by pocketing does not reduce the stock you buy and does add cutting time. Weight reduction pays off in the application, not in the purchase price.
Related reading
Send a drawing for a quote and DFM review.
STEP, IGES, X_T, PDF or DWG up to 50 MB. Tell us quantity and finish, and the quote comes back with tolerance and process notes, not just a price.