Export Carton Planning for Pet Carrier Orders

Pet carrier production desk · Updated 2026-10-07 · 15 min read

Every custom dog carrier we develop is quoted against the same three variables - fabric weight, hardware grade and print method - before the first sample is cut.

Cube is the constraint you will hit first

Every logistics conversation about pet products eventually arrives at the same discovery, usually expressed with some frustration: the container was full and nowhere near its weight limit. Understanding why that happens and what to do about it is most of carton planning.

A 40-foot high-cube container offers around 68 cubic metres of usable internal volume and typically allows something in the region of 26 tonnes of payload. Run those two numbers against a realistic mid-size carrier — say 1.6 kg per unit including its polybag — and the arithmetic is stark. A pack plan achieving 5,112 units puts 8,200 kg inside the container, which is about 32 percent of permitted payload. The volume limit arrives at roughly a third of the weight limit, three times over.

That asymmetry has an important consequence for design decisions. It means grams saved on components are worth very little in freight terms, while cubic centimetres saved are worth a great deal. A board specification that reduces carton dimensions by 20 mm in every direction reduces cube per unit by around 4 percent and increases units per container by the same proportion — which is worth far more than any comparable saving in component cost.

Air freight works on the same principle expressed more brutally. Chargeable weight takes whichever is greater between actual mass and volumetric mass, commonly derived by dividing volume in cubic centimetres by 6,000. Our worked carton above converts to 15.6 kg of volumetric mass against an actual 13.9 kg, so it is already volume-driven on the air line, and the IATA cargo rules most forwarders bill against make no allowance for how much of that volume was air.

The practical instruction to take from this is simple and worth repeating to whoever specifies product rather than freight: if you can remove one centimetre from the packed thickness of the unit, you have done more for landed cost than removing a hundred grams from it.

Reading volumetric divisors carefully

The 6,000 divisor is conventional rather than universal, and some carriers and lanes use 5,000, which increases volumetric mass by a fifth at a stroke. Where a programme regularly moves air, confirm the divisor with the forwarder rather than assuming, because it changes which pack plan is optimal — sometimes decisively, since a plan that is slightly larger in cube but lower in actual mass can flip under a different divisor.

Sea freight quoted in revenue tonnes uses whichever is greater between measurement in cubic metres and weight in tonnes, so the same logic applies with different units. Our example carton at 0.0936 cubic metres against 0.0139 tonnes is quoted on measurement every time.

Units per carton: the trade nobody writes down

Choosing how many units go in a carton looks like a pure cost question — fewer cartons, less board, lower cost per unit. It is that, and it is also three other constraints that pull in the opposite direction.

Handling mass. The more units per carton, the heavier the carton, and somewhere between 15 and 23 kg sits the point at which manual handling guidance and simple human willingness diverge. Our own ceiling is 15 kg gross, which is conservative, defensible against most manual-handling guidance, and — practically — the figure above which warehousemen start dropping boxes rather than lifting them.

Damage. A carton holding twenty units that is dropped has twenty damaged units. A carton holding six has six. Expected damage cost scales with carton contents even though carton count falls, and beyond roughly twelve units per carton the two effects cancel.

Stacking. Taller cartons make taller pallet layers, and the number of layers fits inside the height available above the pallet rather than inside the container height. A plan that looks excellent on cube can be unshippable because it produces a stack 1.9 metres tall including the pallet.

Distribution reality. Many brands do not ship single-SKU full containers. They ship mixed orders, they split consignments between destinations, and they replenish individual sizes. A plan built entirely around maximising units per carton assumes a homogeneity of ordering that most distribution networks do not have, which is why our default recommendation is usually one or two sizes below the mathematical optimum.

The mixed-order argument for middle-sized cartons

A programme with four sizes and three colourways has twelve active SKUs, and an order of 3,000 pieces may well be spread across nine of them. Pack that into sixteen-unit cartons and nine SKUs become cartons holding partial packs, each with wasted air and each requiring individual marking.

Pack it into eight-unit cartons and the same order produces twice as many cartons, each fuller, with far less dead volume inside them. The total cube shipped frequently comes out lower for the middle-sized plan despite what the single-SKU arithmetic suggests, simply because the larger plan cannot be filled.

The review we recommend is to model both plans against last year's actual order distribution rather than against a theoretical single-SKU container. It takes an hour with a spreadsheet and it regularly reverses the conclusion.

Board specification: ECT, burst, and why double-wall is usually wrong

Board is specified two ways and confusing them costs money. Edge-crush test, quoted in kilonewtons per metre, measures the compressive strength of the board's edge and predicts how a stacked box will behave. Burst strength, quoted in kilopascals, measures resistance to puncture and predicts how a box survives being handled roughly. Modern corrugated is specified on ECT, and any supplier quoting only burst is quoting the less useful number.

For a standard pet carrier carton stacked on a pallet, single-wall board at 32 ECT is adequate. It is what we specify by default. Where cartons are stacked three or more layers high on a pallet, or where the destination has a long dwell in humid conditions, 44 ECT is the correct move and costs perhaps 12 to 15 percent more.

Double-wall board is almost never the right answer for this product, despite being specified surprisingly often. It doubles board weight and cost to solve a problem — low stack strength — that is more cheaply addressed by reducing stack height, adding a pallet layer-divider sheet, or improving pallet loading discipline. There are exceptions: very large single-unit cartons for XXL formats genuinely benefit, and any plan where cartons sit directly on the container floor for a long voyage should be reconsidered anyway.

Then specify the closure. Hot-melt adhesive is standard and adequate; gummed tape is cleaner and slower; stitching is stronger and is the correct choice where cartons will bear heavy stack loads or where the closure must survive tampering checks. Whatever is chosen, the top-load compression figure has to be verified against the actual stack rather than assumed from the board grade.

What humidity does to a specification that looked fine

Corrugated absorbs moisture and loses strength doing so. Board that tested comfortably at 32 ECT at ambient conditions can behave like 24 ECT after three weeks in a container holding humid air, and the boxes at the bottom of the stack then deform — usually slowly enough that nobody notices until unloading.

Three controls are worth specifying. Require board with a moisture-resistant adhesive where the route crosses humid climates. Keep cartons off the container floor and away from the walls using pallets and slip sheets rather than relying on dunnage improvised at loading. And reduce internal moisture load at source, which is the subject of the protection section below.

Pallet patterns and the 1.2 by 1.0 problem

Pallet efficiency is where a good carton plan is either realised or lost, because the difference between a well-tiled carton footprint and a poorly tiled one is frequently 15 percent of the pallet area.

Worked on the dominant export footprint, 1,200 by 1,000 mm, our 600 by 400 mm reference carton tiles four per layer with 100 mm left unused along one edge. Rotating alternate layers in a pinwheel pattern recovers some of that: 600 along 1,200 gives two, 400 along 1,000 gives two, which is the same four — but combining one 600-length and one 400-length along the 1,000 mm axis with two 600-lengths along the other yields five per layer with better coverage, at the cost of a weaker interlock.

The rule to hold regardless of pattern is zero overhang. Cartons projecting beyond the pallet edge get struck by fork tines, crushed against container walls, and deformed in exactly the place damage is most expensive — the corners where contents sit. Any pattern leaving overhang should be rejected rather than compensated with stretch wrap.

Stack height then binds. Pallet plus cartons should stay below roughly 1,600 mm for most container work, which gives three layers of a 390 mm carton on a 140 mm pallet with room to spare, and only two layers at 520 mm. This is another reason the eight-unit carton beats the twelve-unit one: it permits an extra pallet layer, and that layer is worth more than the board saved by going larger.

For wood pallets, the export heat-treatment requirement is not optional and not cosmetic, and ISPM 15 compliance with a legible stamp should be checked at loading rather than discovered at customs. The programme is administered in the United States through the U.S. Department of Agriculture, and a single unstamped pallet in a consignment can delay an entire container.

Internal protection: polybags, moisture, and what fails in transit

Whatever happens outside the carton matters less to a customer than what happened inside it, and most in-transit complaints trace to one of four causes.

Abrasion between units. Units touching each other in a carton abrade during a long voyage, and hardware is usually the aggressor: a clip or a buckle rubbing against a neighbour's face fabric for six weeks leaves a visible mark. Unit-level polybags solve this almost entirely for very little cost, and bulk polybag specification is covered in its own note. Where sustainability commitments rule out polybag, a paper interleave works nearly as well.

Moisture. Textiles ship containing more moisture than most people assume, and in a sealed carton crossing a temperature gradient that moisture condenses. Add silica gel at roughly 10 grams per cubic metre of carton volume — around 1 to 2 grams per unit for most formats — specify it as indicating gel rather than standard white so the receiving warehouse can see whether it was breached, and require it loose rather than in a sachet that will be mistaken for a desiccant packet to be thrown away.

Crushing. Units at the bottom of a carton, and cartons at the bottom of a pallet, carry everything above them. Where unit construction includes padding or structured panels, the carton must be packed so the load is carried on the stiff axis rather than through the padded parts — otherwise the units arrive permanently compressed even though nothing was visibly damaged.

Odour. New textile product sealed in a carton for six weeks arrives smelling of itself. This generates more first-week complaints than any genuine defect, and it is addressed by packaging and ventilation rather than by changing the product: adequate dry-down time before packing, a vented carton, and in some cases a charcoal sachet.

Vent holes versus carton strength

Venting helps odour and helps moisture, and both vents and hand-holes reduce box compression strength because they interrupt the walls carrying the stack load. The compromise worth knowing: place vents in the upper third of the side panels rather than centrally, keep total open area below about 4 percent of panel surface, and never place a hand-hole in a corner, which is where virtually all compressive load travels.

A single centrally placed hand-hole can reduce predicted box compression by 10 to 15 percent. Two corner hand-holes can reduce it by considerably more. Both are entirely avoidable with thirty seconds of thought at specification stage.

Carton plan working table

The matrix compares six plans for the same product: a mid-size folding carrier that packs to 480 x 300 x 95 mm before any individual carton is considered. Cartons per container assume a 40-foot high-cube at 68 cubic metres of usable space and 88 percent loading efficiency, both standard working figures.

Read the units-per-container column and the carton cost column together, because they do not agree. Going from eight units to sixteen improves shipped units by 2.7 percent but involves cartons over 27 kg, which is above manual handling guidance more or less everywhere and above what a receiving warehouse will tolerate quietly. Going from eight to twelve gains 2.6 percent and costs compliance with our own 15 kg ceiling.

Meanwhile the move that matters is the top of the table. Individual cartons — even the folded shipper at its excellent 13.7 litres — carry a carton for every unit, and paying a full carton per unit costs more than the 2 or 3 percent of extra volume the master carton would save. The same logic applies to retail-ready formats, which are discussed in the companion note on retail-ready packaging and which require the master-carton question to be answered differently.

Carton marking, documentation, and the receiving check

A correctly marked carton is one a warehouse can process without opening or asking anyone anything. Most failures in this area are not compliance failures; they are friction failures that cost someone twenty minutes per carton for several hours.

Minimum marks on all faces that could be presented upward: consignee mark, destination port, carton number expressed as carton x of y, gross and net mass, and external dimensions in centimetres. Add origin country marking in accordance with the destination's rules, since several markets require it on the carton rather than only on documents, and errors here are among the most expensive clerical mistakes in shipping.

Where the consignment uses mixed SKUs, mark contents by SKU and quantity on the carton rather than relying on the packing list alone. Mixed cartons without external identification are opened, counted, resealed, and eventually short-shipped — every step of which is avoidable.

Packing list structure deserves a note because it interacts directly with how brands plan their orders. Each carton listed line by line against SKU, with carton numbers running sequentially and a per-SKU total, allows a receiving warehouse to reconcile without recounting our work. We issue this as standard and maintain it against our ISO 9001 documentation.

Finally, photograph the first carton of every SKU before sealing and attach the images to the pre-shipment report. When a discrepancy is raised three weeks later on the other side of the world, a photograph of what actually left resolves it in minutes rather than days.

One line saves disproportionate trouble: put a contactable reference for the shipment on the two largest faces, order number plus a single responsible email. When a consignment goes astray, that is what gets it found.

Damage in transit: compression, drops, and the AQL 2.5 handover

Carton planning is finished only once the plan has been tested, and there are two tests that matter.

Box compression test. Empty cartons are compressed between platens until failure, giving a figure against which the actual stack load can be compared. The acceptance rule we use is that predicted top-to-bottom compression must exceed the load the bottom carton carries, with a safety factor between 2.5 and 3.0 for long voyages and approaching 5.0 where cartons sit in ambient humidity for weeks. Test methods for this and for drop resistance are drawn from those published by ASTM International, whose standard practices our laboratories follow so that results are comparable across programmes.

Drop and vibration. A filled carton dropped from a realistic height — 600 to 800 mm in normal distribution, less where palletised throughout — followed by a short vibration run, exposes closure failures and internal movement that a compression test will never show. Where damage occurs, note whether it happened at the closure, at a corner, or through an internal void, because those three have entirely different remedies.

At release we inspect to AQL 2.5 general inspection level II for the product itself, and separately verify the packing operation: carton count and sequence, correct head carton for each SKU, correct marks on all sides, correct desiccant quantity, no overhang on any pallet, and stretch wrap applied to the pallet rather than trapping individual cartons.

The carton plan should also be reviewed if the product changes. Reducing packed thickness by removing individual cartons, altering a fold geometry, or switching to a nested pack are all changes that invalidate the plan, and they are exactly the changes most often made without telling anyone in logistics. Our standing instruction to clients is to re-run the plan whenever the packed unit changes by more than 10 mm in any dimension, and our own work is coordinated from an SGS-verified production base audited to BSCI and ISO 9001, running 7 lines, 137 people, 149 machines, and monthly output of 200,000 pieces.

Standard terms: MOQ 500 per colourway, samples in 6-10 working days with carton data supplied alongside them, bulk production 35-50 days after approval. Payment is T/T with a 30 percent deposit and a 70 percent balance before shipment, quoted FOB Xiamen. For any programme where the packed dimensions are still in flux, the place to start is our companion piece on flat-pack patterns for freight, since folding decisions and carton decisions are one decision made twice.