What you pay for fasteners is far less than what fasteners cost you

The 15-85 rule, and where the arithmetic leads

The formulation comes from Bossard, the Swiss fastener and C-parts management company, and it is blunt: the fasteners themselves are about 15% of the total cost. The other 85% comes from engineering, purchasing, testing, inventory, assembly and logistics.

Bossard also states that savings of more than 50% are achievable in logistics and engineering. Put those two numbers side by side and the direction is clear enough.

Another cost that never appears on the quotation, and who is asking about it: when customers ask for carbon data.

Where the 85% actually goes

“Indirect cost” is too abstract. What actually happens is this, and every line needs people, time and space:

Why the cheapest parts carry the highest management cost

Because management cost has nothing to do with unit price. It follows line-item count.

In ABC analysis, C-parts are the lowest-value class — fasteners, washers, retaining rings and similar. But Bossard's figure is that 65–75% of line items are typically C-parts.

On a 200-line BOM with 140 C-parts, those 140 may contribute under a tenth of the purchase value — and around seven-tenths of the order count, the receipt count, and the counting workload.

Which is why negotiating screws cheaper never feels like it does anything. You are moving the value; the problem is the frequency.

The asymmetry of a stockout

C-parts have one more awkward property: their failure cost is completely out of proportion to their value.

A screw worth a few cents runs out and the whole line stops. What is being assembled on that line may be a cabinet, a machine or a system worth hundreds of thousands — halted by a part number barely worth tracking in your ERP.

So C-parts inventory policy tends to collapse into one of two extremes:

• Hold too much — because nobody dares run out. Shelves fill with low-value parts, capital is tied up, dead stock accumulates.
• Hold too little — because inventory is being controlled. Result: rush orders, air freight, overtime, sometimes a stopped line.

Both extremes are expensive, and neither appears in the unit price of a screw. It is also why vendor managed inventory and scheduled replenishment work particularly well in this category: they address frequency and risk, not unit price. How those arrangements actually work is in VMI and consignment.

Why this is getting harder

High-density computing racks are amplifying the problem. Average data centre rack power density rose from 16 kW to 27 kW within a year, AI racks are already configured above 100 kW, and peak projections reach 1 MW.

Power goes up and the mechanical design follows: heavier structural parts, thicker sheet metal, more complex thermal and power distribution structures. For whoever assembles those racks, three things happen at once:

• More line items — new mechanical designs bring new specifications, and the C-parts part-number count only grows
• Faster revisions — compute platforms advance generation by generation, mechanics follow, and old part numbers become dead stock before they are consumed
• More expensive stoppages — the higher the value per rack, the more alarming the opportunity cost of one missing screw

Six questions to start with

No system needs implementing first. Take stock with these six. The answers are usually surprising — particularly the second and the fifth.

1. How many line items on your BOM are C-parts? What percentage of total line items is that?
2. What share of your purchase value do they represent, and what share of your order count? (Those two numbers are usually far apart.)
3. How many suppliers do you maintain for them?
4. From a C-parts shortage being noticed to stock arriving, how long on average?
5. In the past year, how many times did a low-value part shortage affect output, and for how long each time?
6. What is the value of C-parts dead stock in the warehouse? Is anyone tracking it?

This is not one of the six steps. It shows up across them, or after assembly. Where the decisions that lead here were made is in specifying a screw, which sets out the order and why doing it out of order is rework.

Common questions

What are C-parts, and how do they differ from A-parts and B-parts?

This is the ABC analysis classification, dividing parts into three tiers by value: A-parts highest, B-parts middle, C-parts lowest. Fasteners, washers and retaining rings usually fall into C-parts. The purpose of the classification is to decide how each tier should be managed — high-value parts justify item-by-item calculation, while low-value items that exist in very large numbers should be handled in bulk through a process.

Where does the 15-85 rule come from?

It comes from the total cost of ownership framework put forward by Bossard, the Swiss fastener and C-parts management company. Their statement is that fasteners are around 15 per cent of total cost, with the remaining 85 per cent arising from engineering, purchasing, testing, inventory, assembly and logistics, and that savings of more than 50 per cent are achievable in logistics and engineering. It is a widely cited estimating framework, and the actual proportions vary with industry, product complexity and internal process, so it is worth checking against your own numbers rather than adopting it directly.

Does this mean price comparison is pointless?

No. Comparing prices still needs doing; the point is knowing its ceiling. Price negotiation moves the 15 per cent and usually only by a few percentage points a year, while the way the parts are managed moves the 85 per cent, where the room is far larger. The pragmatic approach is to do both while rebalancing the effort. If a purchasing team spends nine-tenths of its time negotiating and one-tenth on process, that ratio is the inverse of the cost structure.

Does this apply to a small company with fewer line items?

With fewer line items the absolute impact is smaller, but the structure is the same: a handful of low-value items still generate most of the order and receipt transactions. The difference is that a small company usually has no dedicated warehouse staff, so this work is absorbed by engineers or sales people — which is more expensive, because it consumes higher-cost hours. Take stock with the six questions above before deciding whether to change anything.

References

• Bossard America — Total Cost of Ownership (TCO) in Fastening (the original source of the 15-85 rule and the achievable savings)
• Bossard America — What Is C-Parts Management (C-parts definition, ABC analysis, and the 65–75% line-item share)

The proportions vary by industry, product complexity and internal process. Check them against your own numbers rather than adopting them directly.