# Reading a Plant's Cost Sheet: Direct, Indirect, and Overhead in COGS
A single plastic bracket rolls off an injection-molding press every 12 seconds. It costs 8 cents in resin. So why does the finance team argue about whether that bracket is profitable?
Because the resin is the easy part. Everything else (the machine time, the mold wear, the plant manager's salary, the electricity) has to be attached to that bracket somehow. How you attach it changes the reported cost, and therefore the margin. Same part, same press, different number on the P&L.
This lesson pulls apart a plant cost sheet so you can read one with confidence.
COGS (Cost of Goods Sold) is the total cost to produce what you sold. In a manufacturing plant it splits into three buckets.
Materials that physically become the product, traceable to a specific unit.
For our bracket: the polypropylene resin, plus any masterbatch (concentrated color pellets mixed into the resin). If the part needs a metal insert molded in, that is direct material too.
Rule of thumb: if you can point at it in the finished part and count it, it is direct material.
Wages for workers whose time attaches to specific units.
In modern injection molding this bucket is smaller than people expect. The presses run automatically. Direct labor is often just the operator loading inserts or the technician staging a mold. On a highly automated line, direct labor might be a few cents per part.
Everything else needed to run the plant that you cannot trace cleanly to one unit. Also called manufacturing overhead or burden.
Examples in a molding plant:
This bucket is usually the largest and the trickiest, because it has to be *allocated* to parts rather than traced.
> Quick distinction: indirect materials and indirect labor are real costs that support production but do not attach to a single unit. A tube of mold release used across thousands of shots is indirect. The resin in one part is direct. Both are real cash out the door.
Here is a simplified per-part cost sheet for our bracket. The numbers are illustrative, chosen to make the mechanics clear, not real market prices.
| Cost element | Per part | Type |
|---|---|---|
| Polypropylene resin | $0.080 | Direct material |
| Colorant (masterbatch) | $0.010 | Direct material |
| Operator labor | $0.020 | Direct labor |
| Prime cost | $0.110 | |
| Allocated overhead | $0.090 | Factory overhead |
| Total factory cost (COGS) | $0.200 | |
Prime cost is direct materials plus direct labor: the traceable core. Here it is 11 cents. Overhead nearly doubles it.
Notice the leverage. If the bracket sells for 25 cents, the margin looks fine at prime cost (14 cents) but thin after overhead (5 cents). The entire profitability judgment lives in that 9 cents of allocated overhead. So the allocation method is not accounting trivia. It decides the answer.
You cannot trace the plant manager's salary to a single bracket. So you pick a driver, a measurable activity, and spread overhead across parts in proportion to it.
The overhead rate is total budgeted overhead divided by total budgeted driver units.
Say the plant budgets $2,000,000 of overhead and expects 40,000 machine hours next year:
$2,000,000 / 40,000 = $50 per machine hour.
If the bracket runs 4 cavities per shot at 12 seconds per shot, the press makes 4 parts every 12 seconds, or 1,200 parts per machine hour. Overhead per part:
$50 / 1,200 = about $0.042 per part.
Change the driver and the number moves. That is the whole point of the next section.
Watch what happens to our bracket under two choices.
Choice A: allocate overhead by machine hours. Molding is automated and press-heavy, so this tracks reality. The bracket runs fast in a 4-cavity mold, so it absorbs relatively little overhead per part. Overhead lands around 4 to 6 cents. Margin looks healthy.
Choice B: allocate overhead by direct labor cost. The bracket needs an operator to load a metal insert, so it has higher-than-average labor. Under labor-based allocation it now soaks up a disproportionate share of overhead, maybe 12 cents. Suddenly the bracket looks like a loser.
Nothing about the physical part changed. Only the accounting choice did. A plant using labor-based allocation might wrongly discontinue this bracket, or overprice it and lose the bid.
This is exactly the distortion that Activity-Based Costing (ABC) was designed to fix. ABC assigns overhead using multiple drivers matched to what actually causes cost (machine setups, mold changeovers, inspections) instead of one blunt driver. For a plain-English primer, see Investopedia's overview of activity-based costing.
🎬 [VIDEO: "Activity Based Costing Explained" — youtube.com — a clear walkthrough of how ABC reassigns overhead across products and why traditional single-driver costing misleads]
There is a second accounting choice that moves margins: what counts as a product cost versus a period cost.
Under absorption costing (required for external financial statements under GAAP and IFRS), *all* factory overhead, both variable and fixed, gets loaded into the cost of each unit. Fixed overhead like press depreciation sits in inventory until the part sells.
Consequence: if you produce more than you sell, some fixed overhead stays parked in inventory on the balance sheet, and reported profit rises even though nothing sold. Overproduction can flatter earnings.
Under variable costing (used internally for decisions), only variable costs (resin, colorant, operator labor, variable utilities) attach to the unit. Fixed overhead is expensed in full each period.
Variable costing gives a cleaner contribution margin (price minus variable cost), which is what you actually want when deciding whether to take a rush order or drop a part. It cannot be used for published statements, but it prevents the "make more to look profitable" trap.
For our bracket, variable cost might be 14 cents. Any price above 14 cents contributes to covering fixed overhead. That is a very different, and more useful, decision lens than the 20-cent absorption cost.
Knowledge check
1. What is the defining characteristic that separates factory overhead from direct materials and direct labor?
2. A supervisor's salary in a molding plant is classified as indirect labor and lands in overhead, while an operator loading metal inserts is direct labor. What principle explains this difference?
3. The lesson notes that the same bracket on the same press can show a different cost on the P&L. What is the primary reason for this?
4. Why does the lesson describe direct labor as a smaller bucket than people expect in modern injection molding?
5. Select ALL correct answers. Which of the following would be classified as direct materials for the plastic bracket?
Select all the correct answers.
6. Select ALL correct answers. Which costs belong in factory overhead in a molding plant?
Select all the correct answers.
When someone hands you a plant cost sheet, run this checklist.
1. What is the allocation driver? If it is labor hours in an automated plant, be skeptical of every margin on the page.
2. Is the overhead rate current? Rates set on last year's volume go stale. If actual machine hours came in far below budget, the rate was too low and parts were undercosted (this gap is called an overhead variance).
3. Absorption or variable? Know which you are looking at before you compare margins across parts or against a competitor's quote.
4. What is fixed vs. variable inside overhead? For pricing a marginal order, only variable cost matters. For long-run pricing, you must recover fixed overhead too.
5. Are setups and changeovers captured? A part run in short batches triggers frequent mold changes. Single-driver costing hides that; ABC surfaces it.
A common mistake: a plant sees a part with thin absorption margin, drops it, and then discovers the fixed overhead it was covering (that press depreciation) does not disappear. It just reallocates onto the remaining parts, making *them* look worse. Cutting a part that has positive contribution margin can shrink total profit. Read the contribution margin before you cut.