Weight Classes Inside a Pet Carrier Range

Pet carrier production desk · Updated 2026-10-07 · 14 min read

Private label pet bags and carrier programmes share one packaging standard here, so a mixed order does not add a second carton size or handling charge.

Two Different Numbers Live on Every Box: Size Class and Load Class

Almost every range document we review carries a single mass figure per model, used for two jobs that have nothing to do with each other. One job is fit: how big must the interior be for this animal to stand, turn and lie down? The other is strength: how much mass must this structure survive when a nervous animal throws itself sideways at the moment the owner lifts it off the ground?

Those are genuinely different questions and they frequently produce different answers. A tall, lean six-kilogram animal needs decent interior height but generates modest dynamic load. A compact, dense eleven-kilogram animal needs less interior volume and produces far greater peak stress at exactly the moment of lift. Building one united spec to cover both cases either over-engineers the small sizes into a cost band nobody wants, or leaves the dense-animal cases dangerously marginal.

So keep them apart in the documentation. Write the fit dimensions into the size chart and the structural figures into the load class table, and cross-reference the two rather than merging them. This division also makes future negotiation easier: adding a size to the ladder is then a modest incremental cost, while moving a model up one load class is correctly recognised as real re-engineering.

The interface between the two is worth stating explicitly, because it is where mistakes happen: the load class governs nothing about how comfortable the animal is, and the size class governs nothing about whether the bag survives. A model can comfortably fit an animal it is not engineered to carry, which is the most dangerous combination on a shelf.

For background on the fit side of the split, our companion article on breed weight band sizing goes into interior dimension derivation in detail. What follows here is the structural side.

Tracing Where the Load Actually Goes

Before choosing construction, trace the force. There is no substitute for doing this slowly once per design, because it reliably reveals that the money is being spent on the wrong reinforcement.

The primary path is simple: animal resting mass enters through the pad, spreads into the base panel, and leaves upward through the base-to-gusset seams and the handle or strap anchorage. In most soft carriers this path is well understood and rarely fails first. The problems come from the secondary paths.

Path two is dynamic vertical load. Every lift involves acceleration, every step taken while carrying adds a sinusoidal component, and a startled animal moving inside the bag can momentarily increase apparent load several-fold. This is why the static three-times safety factor exists, and why the weakest point is nearly always the strap anchor rather than the base.

Path three is lateral thrust. An animal pressing against one side - particularly when turning - pushes the wall outward and tries to peel the base seam open at a low angle, which is the exact loading mode that stitch lines resist least well. This path is why upper-class designs use a continuous webbing cradle under the floor rather than relying on the seam alone.

Path four is point loading from claws. A claw concentrates force onto a tiny area of the floor or lower wall, and in soft constructions it can work into a seam line and gradually open it. Reinforcing the lower 40-60 mm of every vertical seam, or specifying a base board that extends up the wall a little, addresses this cheaply.

Path five is the carry point itself. Whether the owner lifts by a shoulder strap, a top handle, or two the sacking handle gripped in one hand, each scenario concentrates the whole system load into one or two anchorage points. Conduct your test to the worst of these, not the average.

Path six is improper handling: the bag being dragged, set down hard, wedged into a car footwell, or in some cases picked up single-handed by one panel. No specification can eliminate abuse, but deciding in advance which abuses you will design against is much better than discovering which ones you failed.

Choosing a Safety Factor, and Why Three Is the Working Number

A safety factor converts a rated animal mass into a design test load. Ours is three times, applied to static testing of structural seams and anchorages, and there is a specific technical justification behind choosing that value rather than something larger or smaller.

The dynamic share is the largest contributor. Walking while carrying a loaded bag routinely adds 1.5 to 2 times the static load at the low point of each stride, and a sudden movement by the animal adds further. Two covers normal handling comfortably. A further margin covers material variability - the difference between the mean and the low end of a batch of webbing tensile results - plus degradation over the product's service life, since UV and abrasion reduce both textile strength and seam strength gradually.

Going higher than three has a real cost. Rigidity increases, comfort decreases, the bag gets heavier, and the added weight increases the very load being designed against. A heavy, over-built carrier also retails worse because the tare weight figures in the specifications immediately signal bulk to anyone comparing two products side by side.

Going lower than three is where programmes get into trouble. It is tempting for costing reasons to specify the difference away, and nothing bad happens for the first nine months. Then a photo of a broken strap posted to a retail review appears where every future customer can see it.

The critical caveat: safety factor applies to engineering validation, not to labelling. The test load is three times rated mass; the number printed on the hang tag is rated mass. Never allow a test figure to migrate onto consumer-facing material, since doing so invites a claim neither the construction nor the documentation actually supports.

Independent assessment programmes exist for this category of product and are worth reviewing before writing claims, since what a third party is prepared to certify is a reasonable indicator of what it is prudent to state. Center for Pet Safety publishes its protocols openly, and even if certification is not pursued, reading them improves how any team writes its own specification.

The Weight Class Table: What Changes Between Tiers

Four classes cover essentially every pet carrier programme we have supported, from soft cat totes to crate-style travel bags. The jump from class two to class three is the interesting one, because that is where a soft bag stops being purely textile and starts needing genuine structural members.

Read the anchorage column carefully, because it is where the money genuinely goes. Moving from a sewn return to a continuous webbing cradle is a construction change, not a material upgrade: the strap continues around the base, so the animal's mass is carried by a tension member rather than by a shear-loaded stitch line. It costs one continuous length of webbing instead of two cut ends, and it eliminates the most common heavy-class failure with no weight penalty.

The floor column has its own trap. Board deflection is felt long before board failure, and a floor that visibly bows under a twelve-kilogram animal reads as cheap even though it never breaks. Specifying deflection under load rather than simply board thickness is the more honest control, and the two materials behave quite differently in that respect - our baseboard material comparison sets out the practical trade-offs across PP board, honeycomb and laminate options.

Note also what the table does not say: it does not pair classes with sizes. A compact model for a dense animal can legitimately be a size small with a Class C structure. Being able to specify those independently is the whole point of separating the two number sets.

Where stitch density meets weight class

Load class raises stitch density requirements only at anchorage points, not across the whole bag. A Class D strap return wants 10-12 SPI in Tex 70 thread with a bar tack; nothing else in the bag changes stitch specification because the animal weighs more. Programmes that raise stitch density globally across a heavier model pay more to gain nothing.

Three Tests That Predict Real Failure Better Than Any Certificate

Certificates verify that a construction met a standard once. What predicts field performance is a test matrix run on the actual production representative unit, repeated on every significant construction change. Three protocols do most of the work.

The first is static proof load. Load the unit with ballast to three times rated animal mass, suspend or support it by the primary carry point, and hold for a defined period - five minutes is workable, twenty-four hours is more informative for creep. Acceptance is no thread breakage, no anchorage movement beyond a stated allowance, and no permanent deformation once unloaded. Photograph every fixture setup, because most arguments are about how it was loaded rather than the result.

The second is loaded drop. Fill to rated animal mass with a distributed ballast bag rather than a point mass, lift to 0.5 m, and release onto a hard floor in six orientations: base, each long side, each end and the top with the handle uppermost. The handle-up case is deliberately included because nobody lifts a bag that way deliberately, but plenty do accidentally. Failure at this stage usually reveals a board that cracks rather than a seam that tears.

The third is cyclic actuation, which catches the failures the first two miss. Cycle the strap or handle attachment 5,000 times at rated load with a low-frequency lift-and-set motion, then re-run the static proof and compare results. A fall-off in residual strength indicates progressive damage - usually webbing abrading against a return edge or stitch abrading against a hardware slot - which no single-event test will ever show.

Add cold-condition testing if any part of your distribution enters a genuinely cold market, because polypropylene board becomes markedly more brittle below freezing and a board that survives a drop at twenty degrees may crack at minus five. This is a common and entirely avoidable seasonal return.

Finally, standardised strength methods matter for comparability across seasons and partners. Test procedures for tensile behaviour of textiles and webbings are published by ASTM International, and referencing a method rather than a procedure described in prose is what makes a test result reproducible eighteen months later by somebody who was not in the room.

Building Three Commercial Tiers Around Four Load Classes

The engineering question is how much load; the commercial question is how many tiers to sell, to whom, at what price structure. These interact directly, because construction class is the most legible expression of value a customer can feel without being told.

A three-tier range maps cleanly onto the four classes. The entry tier sits in Class A and B, kept deliberately simple: one or two sizes, standard webbing, no frame, one decoration method. It exists to win shelf presence and first-purchase price points, and importantly, it should be built from the same materials as everything above it so quality perception stays consistent.

The middle tier is where most branded programmes make their margin, and it spans Class B and C. This tier justifies visible structural features that photography supports: wider webbing, a branded pull, a cradle detail visible through a mesh panel, a removable structured pad. Customers will pay a meaningful premium for construction they can see and name.

The top tier covers Class C and D and supports either a genuinely heavy-animal proposition or an aesthetic one built on material rather than pure strength - and it should be honest about which it is. A top tier that claims heavy capability but uses middle-tier anchorage is the fastest route to a damaging review.

Price ladders between tiers are typically 1.4 to 1.8 times step-ups, which is tighter than apparel because component cost does not rise proportionally. Resist compressing tiers closer than that: below roughly 1.3 times, customers cannot tell what they are paying for and default to the cheapest.

Range planning should also fix where each tier will be sold, because channel determines construction requirements more than any other factor. Warehouse-club and mass channels need the carton cube and price point of entry tiers; specialty pet and veterinary channels will carry the reinforced units and expect documentation; outdoor and lifestyle accounts respond to the material-led premium proposition.

What You Are Allowed to Say on Packaging

Load class engineering creates claims, and claims create regulatory exposure. The general principle is simple: state what you have verified, and do not imply certification you have not obtained.

Stating a maximum animal mass is generally acceptable where it is supported by test evidence retained on file. So are construction descriptions: reinforced base, continuous webbing cradle, removable structured pad, and so on. Both describe what exists and both can be substantiated with a report.

The problems start with performance adjectives. A word like crash-tested implies a specific third-party protocol that neither this product nor any soft carrier is normally submitted to, and it creates a duty of care that no amount of construction actually discharges. Similarly, any implication of veterinary or safety endorsement moves the claim into regulated territory, since transport-restraint products attract formal assessment.

Consumer product safety obligations also apply to general product representation in the markets where you sell. Guidance published by the U.S. Consumer Product Safety Commission is the baseline reference for US-bound programmes, and it is worth having your marketing copy reviewed against the requirement that claims be substantiated before the first print run rather than after.

Keep the care label honest too. If the construction is only tested to line drying, do not print tumble dry. If the pad cover is machine washable but the shell is wipe clean only, say so per component - mixed messaging here is a surprisingly common source of complaints, and unlike strength claims it is entirely within the brand's control.

The safest and most effective format for packaging is a small spec block: rated animal mass, tare weight, principal dimensions, principal materials, and care instructions. It reads as confidence rather than as advertising, and everything in it can be produced on request.

Field Failures by Class, and the Design Change That Fixes Each

Every repeated failure pattern we see has a design-level answer that costs less than the returns it prevents. The following four account for the great majority of heavy-class complaints in soft carriers.

The first is prescribed base sag. The construction uses a board adequate in strength but too thin in stiffness, so the floor bows under load and the animal sits in a shallow hammock. The answer is not a thicker board but a stiffer one - honeycomb or a laminated sandwich resists deflection far better per gram than solid sheet - or alternatively a cradle that takes the load off the floor entirely.

The second is handle-webbing pull-out at the riser. It almost always indicates insufficient return length, not insufficient stitch density. Extending the return past the anchor point and adding an X-box redistributes peel load into shear across a much longer thread path, and it costs a few centimetres of webbing.

The third is board cracking at the fold line, which occurs when a board with a crease is specified in a construction where the crease is also a load line. Either move the crease outside the loaded area or specify a board designed without a folding crease at all, according to whether flat-pack capability is genuinely required.

The fourth is seam tear at the lower corners of vertical panels, driven by claw point loading. Extending the base material up the wall by 40-60 mm so claws contact a continuous surface rather than a sewn join removes this failure almost completely.

The commercial version of this same advice: track returns by failure mode rather than by SKU. A list of thirty broken bags sorted by model number tells you nothing; the same thirty sorted by what actually failed will point at a single pattern correction that ends the problem.

Having worked through these six areas - load paths, safety factor, class table, test matrix, tier planning, and honest claims - the range stops being a set of similar bags in ascending sizes, and becomes a defensible structure with engineering and pricing that reinforce each other.