Manufacturing Cost Accounting

The Only Four Cells a Cost-Per-Part Spreadsheet Needs

Cost-per-part spreadsheets grow columns the way boats grow barnacles. Four inputs decide the answer, and yield is the one that moves it most while getting the least attention.

A cost-per-part model does not get more accurate as it gets longer. Past a certain point it gets less accurate, because each additional input is another assumption nobody revisits, and a model with forty cells has thirty-six places to be quietly wrong. Four inputs decide the answer for most discrete parts: material cost net of scrap credit, cycle time, first-pass yield, and a loaded rate for the resource. Everything else is either a refinement of one of those or a rounding error dressed as rigour.

The formula, and where yield enters

Cost per good part is material plus labor and machine cost for the cycle, all divided by yield. That final division is the step people forget, and it is the one with the most leverage. Work it: a 2-minute cycle at the current manufacturing wage of $30.35/hour (Jul 2026, BLS) with a mid-range 35% burden gives a loaded rate near $41 per hour, so labor content is about $1 per part. Add $2 of material and divide by 92% yield and the cost per sellable part comes out near $3. Drop yield five points and that figure moves more than a ten percent material saving would.

Why more columns make it worse

Every added input needs an owner and a review date, and models rarely get either. The common failure is a sheet with painstaking detail on consumables, tooling inserts, and packaging, sitting next to a cycle time that was measured once during a good run three years ago. Precision on the small terms creates false confidence in a model whose largest term is stale. If you have an afternoon to spend on a cost model, spend it re-measuring cycle time and pulling actual yield from the quality system, not on adding line items.

A cost model is only as good as its worst-maintained input, and that is almost always cycle time, sitting quietly under forty cells of consumable detail.

Overhead is a choice, so make it visible

The loaded rate embeds an allocation decision, and the same part can carry meaningfully different overhead depending on whether you allocate by machine hour, labor hour, or square footage. None of those is wrong, but the choice changes which products look profitable, so it belongs in the open. Put the allocation basis and the burden multiplier in their own labelled cells rather than folding them into a single rate. When someone challenges a quote, being able to show the multiplier and its basis ends the argument in a way an unexplained hourly figure never will.

Use the burden rate calculator to derive a loaded hourly rate from your own general ledger. Build the rate properly

Published 2026-08-08.