Key takeaways
- One damaged panel costs about 1.3 times its piece price once inspection, return freight and an expedited replacement are counted, so a pack that costs 1.6% more of part value has to prevent 1.2 percentage points of damage to pay for itself.
- A pack level that adds ₹55 per panel against a ₹2,750 claim breaks even at a 2.0-point reduction in damage, which is the only number that should decide between two quotations.
- Forty 1.5 mm panels at 7.1 kg put 309 kg on the pallet below, and five pallets high puts 1,236 kg through the bottom cartons, more than a 5-ply carton can carry at 90% relative humidity.
- A 200-unit trial cannot separate a 3% damage rate from a 0.8% one; about 600 units per pack variant are needed before the difference is visible at 80% power.
- Crates reduce damage but rarely win on cost: at 6% of part value they beat a carton only when the stack load or the part size rules the carton out.
Two quotations for the same 1.5 mm enclosure panel arrive with different packs. One is a single-wall carton with corner boards; the other is five-ply with foam interleaving and a VCI liner, and it costs 1.6% of the part value more. The buyer has to decide whether that 1.6% buys anything.
It buys 2.2 points of damage, in this example, which is worth 2.9% of part value. The expensive pack is the cheaper one. Reverse the numbers and the same arithmetic says the cheap pack wins. Packaging cost is not a line item to minimise; it is a trade against a claim rate, and the trade has a crossover point you can calculate before the first shipment.
The sections below build the claim cost, find the break-even, test it against a real stack, and set out the sample size needed to prove the change.
What does one damaged part actually cost?
A damaged part costs more than its piece price. The claim includes the replacement, the inspection time at both ends, the return freight, the expedited outbound freight and the packaging that is scrapped with it.
Take the replacement as 1.00 P, where P is the piece price. On a fabricated panel, the rest of the claim typically runs:
- Goods-in inspection and claim administration: 0.10 P.
- Return freight for the rejected part: 0.08 P.
- Expedited replacement freight, because the customer has a build schedule: 0.12 P.
- Pack materials consumed with the rejected part: 0.02 P.
That is 1.32 P per claim, before anything goes wrong on the customer's line. The multiplier is the number most pack decisions ignore: a pack that costs 1% of part value is competing against a claim that costs 132% of it, so 1% of pack spend buys 0.76 points of damage reduction and is still worth it.
Line stoppage should not be averaged into this number. If a missing panel stops an assembly line for an hour, the cost can exceed a year of pack spend on that part, and an average hides the event. Treat a serial part with a line-side schedule as a separate case with a much higher claim cost, and pack it accordingly.
Where is the break-even between packaging cost and damage?
The break-even is the damage reduction that repays the extra pack cost:
Break-even (percentage points) = (pack cost B − pack cost A) ÷ claim cost
Both terms as a percentage of part value. If pack B costs 1.6% more and a claim costs 132%, the break-even is 1.6 ÷ 132 = 1.2 points. Pack B pays for itself if it takes damage from 3.0% to 1.8% or lower.
Run it as a total cost per unit instead and the same answer appears:
- Pack A: 1.0% of P pack cost plus 3.0% × 132% = 1.0 + 3.96 = 4.96% of P.
- Pack B: 2.6% of P pack cost plus 0.8% × 132% = 2.6 + 1.06 = 3.66% of P.
On 20,000 panels at an illustrative ₹1,000 piece price, that is ₹13 per panel, or ₹2.6 lakh a year. The rupee figure moves with the piece price; the ratio does not, which is why the model is worth building in percentages.
Two sensitivities matter more than the rest:
Claim cost. At 3 P — a serial part with a line-stop risk — the break-even falls to 0.5 points, and almost any pack upgrade pays. At 1.32 P, the pack has to be aimed precisely.
Volume. At 200 panels a year, the fixed cost of a new pack (die, liner, test) dominates and the ratio model understates the decision. At 20,000, the pack material and the damage rate are the only terms that matter.
What do the four pack levels actually cost?
The table puts four levels against the same panel: 1.5 mm CRCA, 0.6 m², 7.1 kg, powder coated on one face. Costs are percentages of part value, and damage rates are the middle of the range seen on fabricated panels shipped by road.
| Pack level | What it adds | Damage rate | Pack cost (% of P) | Total cost (% of P) | Ganesh limit or note |
|---|---|---|---|---|---|
| Bulk stack on a pallet | nothing between parts | 6.0% | 0.3% | 8.2% | only for parts that will be recoated or machined again |
| Carton with corner boards | single-wall carton, 3 mm angle board, kraft interleave | 3.0% | 1.0% | 4.96% | adequate for 1.5 mm panels over short road legs |
| Five-ply carton, foam and VCI | five-ply, EPE corners, foam interleave, VCI liner | 0.8% | 2.6% | 3.66% | the default for coated visible panels |
| Crate | timber crate, internal blocking, bolted base | 0.3% | 6.0% | 6.4% | chosen for load or size, not for damage rate |
The crate row is the useful one. It has the lowest damage rate and the highest total cost, because it is solving a load problem that a carton cannot solve, at a price the damage rate does not repay. Crates are correct when the stack load exceeds the carton's capacity, when the part is too large for a pallet, or when the receiving dock cannot handle cartons.
The bulk row is also instructive. Six per cent damage on a part that is inspected and recoated anyway is not damage; it is handling. A part that will be machined after fabrication does not need face protection, and paying for it is the most common form of over-packing.
Why can more packaging raise the damage rate?
Packaging that is too tight damages the part it protects. Three mechanisms show up repeatedly:
Compression. Foam that is thicker than the gap between parts preloads the stack. A stack of 20 panels with 6 mm foam between them adds 120 mm of compressed foam, and the bottom panel sees the closing force of every layer. The damage is a soft dish in the middle of a face, which the pack was bought to prevent.
Hidden damage. A crate with a sealed liner hides a dent until the customer opens it, often weeks later and often after the parts have been booked in. The damage rate at goods-in looks excellent and the rate at the line looks terrible.
Unpacking time. A pack that takes four minutes to open costs labour at the customer's dock that never appears in the pack price. On 20,000 panels, three extra minutes is 1,000 hours.
The test is not whether the pack protects the part. It is whether the pack protects the part at the lowest total cost, which includes the customer's time.
What does the stack decide, independent of the pack?
Stack height is decided by the load path, not by how many pallets fit under the roof. Work it from the part up.
A 1.5 mm CRCA panel at 0.6 m² weighs 7.1 kg. Forty panels plus a 25 kg pallet is 309 kg. A 1,200 × 1,000 mm pallet holds four cartons at 600 × 500 mm, so each carton in the bottom layer carries one quarter of the pallet above it, 77 kg.
A five-ply carton with an edge crush test of 32 lb/in (5.6 kN/m), a 5 mm caliper and a 4.4 m perimeter has a box compression strength of about 4.88 kN, or 497 kg, from the McKee relation:
BCT = 5.87 × ECT × √(perimeter × caliper)
At a safety factor of 3 for a 60-day stack, one carton carries 166 kg dry. At 90% RH, derated by half, it carries 83 kg. That is a factor of 1.1 on a two-high stack of 77 kg per carton, and a failure at three high.
This is why the pack conversation has to include the stack height and the warehouse humidity. A quotation that specifies a carton without them is specifying half a pack. Where the height is fixed at five pallets, the answer is a pallet collar, a crate or racking, not a thicker carton.
How do you measure the damage rate well enough to decide?
A damage rate is only useful if it is defined, counted and sampled the same way on both sides.
Define damage by consequence, not by appearance: any deviation that affects fit, function, coating integrity or the customer's first operation. A scuff on a face that will be hidden inside an enclosure is not damage; a scuff on a visible powder-coated face is. Write the definition into the pack specification so the customer and the supplier count the same events.
Count at goods-in, per lot, with a photograph and the pallet position. A damage rate that is not tied to a pallet position cannot tell you whether the cause was the pack, the stack or the forklift.
Then size the trial. A 200-unit comparison of 3.0% against 0.8% will produce a number, but the confidence intervals overlap and the result is noise. At 80% power and 95% confidence, detecting a 2.2-point difference needs about 600 units per pack variant. That is 1,200 panels shipped, which is two to four weeks for a mid-volume part and worth planning before the change is made.
Until the trial is complete, use the arithmetic in the previous sections with the damage rates you already have. A decision made on a 1.2-point break-even and a 3.0% observed rate is more defensible than one made on a 200-unit trial that happened to come out well.
What does Ganesh pack, and what it needs from the drawing?
Packing is quoted with the part, because the pack has to suit the part's mass, finish and handling. The range we fabricate runs 0.8–40 mm, which spans a 0.8 mm cover panel at 6.3 kg/m² and a 6 mm base plate at 47.1 kg/m². The first is a two-person lift and the second needs a forklift, a different pallet and a different pack.
The finish decides the interface. A bare or phosphated part can be stacked with interleaving and shipped in a VCI liner. A powder-coated visible panel needs face protection, a foam or kraft interleave and a pack that does not preload the coating. Coated assemblies that leave the plant through the finished coated assemblies route are packed after the cure has cooled, not before, because a film packed warm prints against itself.
What we do not do is run a full ISTA or ASTM D4169 qualification in-house for every part. Where the customer's contract requires a drop, vibration or compression test to a named procedure, it is specified in the quotation as a separate step with its own lead time, rather than assumed.
The plant quality and sustainability hub covers the receiving and inspection side. Packaging for transit covers the materials and the protection mechanics, packaging damage in transit covers how to read a dent at goods-in, and the outdoor powder coating failures guide covers the coating defects that a pack can cause. The custom parts page lists the thickness range that drives the handling plan.
What to put in the pack specification
- Part number, revision and quantity per pack, with the gross mass.
- Maximum stack height in pallets and the warehouse humidity the pack is rated for.
- Support points and the face that must not be loaded.
- Interleave material and thickness, and the faces it goes between.
- VCI or desiccant, with the barrier area it is sized to and the storage period.
- Strapping type, tension and edge protection.
- Label data: part number, revision, quantity, lot, gross mass and handling symbols.
- The damage definition and the acceptance sampling at goods-in.
- The test procedure and level, if the contract requires one, named explicitly.
When a crate or a returnable is the cheap answer
A crate is the cheap answer when the carton is structurally impossible: a 2,500 mm panel, a 900 kg assembly or a stack that has to go five high in a humid warehouse. At that point the crate is not competing on damage rate, it is the only pack that carries the load, and the 6% of part value buys the shipment.
A returnable stillage is the cheap answer when the same part ships to the same dock more than about 30 times a year and the empty stillage comes back on a controlled leg. Below that frequency, the return freight and the pool administration cost more than the cartons, and the stillage sits in the customer's yard.
For everything else, the choice is a carton and a break-even calculation. Send the drawing, the pack quantity, the shipping mode and the damage rate you are seeing to request a quote, and the pack will be quoted against the claim cost rather than against a catalogue.
Frequently asked questions
Is it cheaper to over-pack or to accept some damage?
Compare the extra pack cost with the claim cost times the damage reduction. If a better pack costs 1.6% of part value and a claim costs 132% of part value, it breaks even at 1.2 points of damage reduction. Above that, the better pack is cheaper; below it, the damage is cheaper than the fix.
How much should packaging cost as a share of part value?
For fabricated steel parts, 1–3% of piece price covers cartons, corner protection, interleaving and palletising. Above 5%, the pack is usually compensating for handling damage inside the plant rather than transit. Crates sit outside that band because they solve a load problem.
How many parts do I need to test to prove a packaging change?
About 600 shipped units per variant to detect a two-point difference in damage rate at 80% power. A 100-unit trial will produce a result, but the confidence interval will be wider than the difference you are trying to measure.
Why does a carton that passes a dry test fail in a warehouse?
Corrugated loses roughly half its compression strength at 90% relative humidity. A box tested at 50% RH and rated at 497 kg is carrying 248 kg in a monsoon warehouse, and a three-high stack will crush the bottom layer.
Should coated parts be wrapped in stretch film?
Loosely, and only after the film has cured and cooled. Film applied at more than about 20% elongation pulls the coating against itself and prints the edges. Interleave first, then wrap, and leave a vent so condensation can escape.
When is a returnable pack justified?
When the same part runs for more than about 30 shipments a year to the same dock and the return leg is controlled. Below that, the return freight and the pool administration cost more than the cartons it replaces.
Sources and further reading
- ISO 4180:2009 — Packaging, complete filled transport packages, general rules for compiling performance test schedules · reviewed 3 Nov 2026
- ASTM D4169-22 — Standard practice for performance testing of shipping containers and systems, distribution cycles and assurance levels · reviewed 3 Nov 2026
- ISO 2233 — Packaging, complete filled transport packages, conditioning for testing, including the humidity conditions that derate corrugated · reviewed 3 Nov 2026
Comparing two pack specifications?
Send the drawing, the pack quantity, the shipping mode and the damage rate you are seeing. We will compare the pack cost against the claim cost and quote the pack with the parts.
Request a quote