Sourcing
Landed Cost: What a Part Really Costs
The short answer
Landed cost is the total cost of getting a part into your warehouse, ready to use. It includes the unit price, but also tooling amortization, freight, duties, brokerage, insurance, inspection, packaging, payment and currency fees, and an allowance for defects. Comparing suppliers on unit price alone routinely picks the more expensive option.
Every line item
| Line item | Commonly missed? | Notes |
|---|---|---|
| Unit price | No | The number everyone compares. Often under half the total |
| Tooling, amortized | Sometimes | Divide tool cost by the realistic quantity, not the hoped-for one |
| International freight | No | Air versus sea changes both cost and lead time dramatically |
| Duties and tariffs | Yes | Depends on HS code and country of origin. Verify the code |
| Customs brokerage | Yes | Per-shipment fee, hurts most on small frequent orders |
| Insurance | Yes | Usually small, but the claim process matters more than the premium |
| Inland freight | Yes | Port to warehouse, plus any drayage |
| Incoming inspection | Yes | Real labor, especially on a new supplier |
| Packaging and repack | Yes | Bulk-packed parts often need repacking for your line |
| Payment and FX fees | Yes | Wire fees plus spread. Material on frequent small orders |
| Inventory carrying cost | Yes | Longer lead times force more stock, which is capital sitting still |
| Defect and rework allowance | Yes | A 3% defect rate is a 3% cost increase, plus the labor to find it |
The two that move the number most
Tooling amortization is the first. A $12,000 tool spread across the 500 units you will actually build this year is $24 per part, not the $2 it becomes at 6,000 units. Teams routinely amortize against the optimistic forecast and then discover the real number after the fact. Amortize against the quantity you are confident about.
Inventory carrying cost is the second, and it is the one almost nobody puts on the spreadsheet. A supplier with a twelve-week lead time forces you to hold far more safety stock than one at four weeks. That stock is capital you cannot spend, plus warehouse space, plus obsolescence risk every time you revise the design. On a fast-moving hardware product where revisions are frequent, obsolescence alone can dwarf the unit price difference.
Where the comparison usually goes wrong
Three patterns repeat.
Comparing an overseas unit price against a domestic one without adding freight, duty and brokerage. By the time those land, the gap is often much smaller than it looked, and sometimes reversed.
Ignoring the cost of your own time. A supplier who needs three rounds of clarification on every order is consuming engineering hours that do not appear on any invoice.
Treating lead time as a schedule issue rather than a cost. It is both. Longer lead times mean more inventory, more risk on design changes, and less ability to respond when demand moves.
A practical way to compare
Build one spreadsheet row per supplier with every line item above, and compute cost per usable part delivered. Usable matters: divide by the yield you actually expect, not the quantity shipped.
Then add a column for lead time and one for the revision risk, meaning how much inventory you would be writing off if you changed the design next quarter. Those two columns will not produce a number, but they will frequently change the decision.
Questions to ask before you commit to a supplier
A landed-cost spreadsheet tells you what a part costs today. These questions tell you what it will cost you over the next year, which is usually the number that decides whether the relationship works.
What is the price at half my forecast volume? Forecasts are wrong, and the answer tells you how much of your quote depends on a quantity you may not hit.
Who owns the tooling, and what does it cost to move it? Settle this before the purchase order, not during a dispute.
What is your quoted lead time, and what is your actual lead time over the last six months? These are different numbers and a good supplier will tell you both.
What happens when I change the design? Engineering changes are certain on a young hardware product. The cost and lead time of a revision matters more than the initial price.
What is your defect rate on parts like this, and how do you handle rework? A supplier who says zero either has no data or is not telling you the truth.
Who do I call when something is wrong? A named person who understands your part is worth a real premium over a cheaper quote and a shared inbox.
A worked comparison
The clearest way to see why unit price misleads is to run two quotes all the way to the warehouse. Take 1,000 machined aluminum brackets, quoted by a domestic shop at a higher unit price and an overseas shop at a lower one.
| Line item | Supplier A (domestic) | Supplier B (overseas) |
|---|---|---|
| Unit price | Higher | Substantially lower |
| Tooling and setup, amortized | Lower setup, no tooling | Similar or lower |
| Freight | Ground, low | Air or sea, materially higher |
| Duty and tariffs | None | Applies, rate depends on HS code |
| Customs brokerage | None | Per shipment |
| Incoming inspection | Light | Heavier on a new supplier |
| Lead time | 3 to 4 weeks | 10 to 14 weeks |
| Safety stock implied | Low | High, which is capital held still |
| Cost of a design change | One short cycle | A full pipeline of in-transit stock at risk |
The unit price gap is real, and at high volume it usually still wins. At low and mid volume, freight, duty, brokerage and the inventory a long lead time forces frequently close most of it, and the design-change exposure closes the rest on a product that is still moving.
Where landed cost changes the decision, and where it does not
Running the full calculation is not always worth the effort, and pretending otherwise makes it a ritual instead of a tool.
It changes the decision most when volumes are low or medium, when the product is still revising frequently, when the parts are physically bulky relative to their value, when duty rates are high for the classification, or when your cash position makes carrying inventory genuinely painful. In those situations the cheapest unit price is regularly the most expensive option overall.
It changes the decision least when volumes are high and stable, the design is frozen, the parts are small and dense, and the unit price gap is very large. There, freight and duty become rounding errors against the piece price and the obvious answer is usually the right one.
The practical middle ground: run the full landed calculation once per part family rather than per quote. The ratio between unit price and total landed cost is fairly stable within a family, so once you know that a machined aluminum part from a given region lands at roughly a known multiple of its quoted price, you can compare future quotes quickly without rebuilding the spreadsheet each time.