Total cost of ownership vs piece price

Why piece price misleads

A piece price is a single number on a quotation. It answers one question: what does the supplier charge for one part at the agreed Incoterm? It says nothing about the cost of moving that part, storing it, inspecting it, or recovering from a batch that fails in production.

Two suppliers can quote the same part at very different prices and still deliver the same total cost. One may be close to your plant with a short lead time and a clean quality record. The other may be cheaper per part but slower, further away and more likely to need sorting at goods in. The invoice is the visible part of the cost. The rest sits in overheads, working capital and line downtime, where nobody looks until something goes wrong.

Total cost of ownership, usually shortened to TCO, is the discipline of putting all of those lines in one place before a sourcing decision is made. It does not need to be complicated. It needs to be consistent.

The lines that belong in the model

A workable model starts with the goods value and then adds every cost that changes between the options under review. The exact list depends on the part and the route, but seven groups cover most engineered components bought from overseas or from a domestic supplier. Pick the lines that actually differ between the options, because a line that is identical for every supplier only adds noise.

The groups are not equally visible. Piece price arrives in the quote. Logistics arrives in the freight invoice. Duty arrives later, sometimes after the goods have been sold. Inventory, quality and administration rarely arrive as a single invoice at all, which is why they are the groups most often left out of a comparison.

• Piece price: the unit price at the agreed Incoterm and volume, including the volume tier assumed in the quote.
• Logistics: main freight, insurance, clearance and inland delivery, plus demurrage and expedited freight when they occur.
• Duty and taxes: customs duty, anti-dumping duty and any non-recoverable tax, based on the correct classification and origin.
• Inventory: safety stock, carrying cost and obsolescence, including the cash tied up by a longer lead time.
• Quality: incoming inspection, rejects, sorting, rework, scrap, warranty exposure and the risk of a line stop.
• Administration: purchase orders, supplier management, audits, travel and the engineering time spent chasing documents.
• Tooling: amortised tool cost, maintenance and replacement, together with the terms for ownership and transfer.

Quality and delivery costs are the hard part

Most teams can estimate freight and duty. Fewer can put a number on quality. The temptation is to leave it out because it is uncertain, but leaving it out quietly assumes every supplier performs the same way. That assumption is usually wrong, and it always favours the supplier with the least evidence.

A simple approach is to use the supplier's recent reject rate where one exists, and a conservative estimate for a new supplier until a first article and a few shipments have been measured. Apply that rate to the annual volume, multiply by the cost of a rejected part, and add the labour for sorting and the risk of a line stop.

• Rejects and scrap at the agreed price, not at cost
• Incoming inspection time, including sampling and documentation checks
• Sorting and rework when a batch is found non-conforming
• Expedited freight to protect a production date
• Safety stock needed to cover a longer or less reliable lead time
• Engineering time for drawing clarification and corrective actions
• Warranty and field failure exposure for safety-related parts

An illustrative comparison

The numbers below are illustrative and are not a market rate. They show how a model changes a decision. Supplier A quotes a lower piece price but ships from further away, carries a longer lead time and has a higher reject rate. Supplier B is closer and more expensive per part.

For an annual volume of 50,000 parts, Supplier A lands at a higher cost per good part once freight, duty, inventory and quality are included. The gap is not dramatic, but it is enough to change the award. The point is not that the distant supplier always loses. The point is that the decision should be made on the total, not on the quote.

When piece price should win

Total cost does not mean the lowest total always wins. There are cases where piece price should carry most of the decision. A commodity part with a stable design, several qualified suppliers and a short, reliable route may have almost no difference in the non-price lines. In that case, a detailed model adds work without changing the answer.

Piece price should also dominate when the volume is very low. A prototype order of fifty parts will not carry meaningful inventory or tooling amortisation, and the administrative cost of a full model can exceed the value of the decision. Use a short version of the model and focus on the lines that can actually move.

The judgement is about materiality. If the non-price lines are small relative to the piece price difference, say so and move on. If they are the same order of magnitude, build the model.

Agree the assumptions before the numbers

A total cost model is only as good as its assumptions, and assumptions are where internal arguments start. Put them on one page before the comparison: annual volume, order frequency, lead time, payment terms, Incoterm, duty rate, currency, reject rate and the cost of capital.

Where a number is uncertain, show a range rather than a single figure. If the duty rate is not confirmed, mark it as provisional and state who will confirm it. If the reject rate is an estimate, label it as such. A model that hides its uncertainty invites false confidence.

It also helps to agree who owns each line. Logistics owns freight. Finance owns the cost of capital. Quality owns the reject rate. When each line has an owner, the model becomes a shared document rather than a procurement spreadsheet that everyone disputes at the end.

Keep the model alive

Freight rates, duty, currency and supplier prices all move. A model built for a quote in January may be wrong by June. Date every assumption, keep the source beside each rate, and review the model when a quote is older than the basis behind it.

For a long programme, agree which inputs are indexed and how often they are reviewed. A metal index, a currency clause and a freight review cadence turn a one-off calculation into a commercial agreement that both sides can defend. That is the difference between a sourcing decision and a sourcing relationship, and it is what keeps the model credible when the next quote arrives.