Cat Carrier Return Policy: B2B Guide

Pet carrier production desk · Updated 2026-10-06 · 16 min read

Wholesale pet carrier programmes for cat carrier ranges run 35-50 days after sample approval, shipped FOB Xiamen under T/T 30/70 terms.

Two Instruments: Warranty and Returns Are Not the Same

A warranty and a return policy are routinely conflated, and the conflation costs money because each is applied to the situation the other was designed for. Keeping them separate in the contract is the single most useful thing a buyer can do at the negotiation stage.

A warranty is a promise about conformity over time. It is triggered by a defect, it runs for a period measured in months, and its remedy is repair, replacement or credit. It does not involve the physical return of the goods in most well-designed programmes, because freight dominates the cost of a low-value remedy.

A return policy is a process about goods moving backwards. It is triggered by a decision to send something back, it runs on a window measured in days, and its remedy is a credit or a replacement against received goods. It applies to situations a warranty does not cover — an over-order, a slow-moving SKU, a specification change, a retailer's seasonal reset, or a shipment that failed to meet the agreed inspection standard.

The overlap is the defect return, and it is where the distinction matters most. A defective unit returned within thirty days is both a warranty claim and a return; the warranty governs whether it is paid for, and the return policy governs how it travels and how it is credited. Writing one clause to cover both produces a document that answers neither question well.

Consumer returns are a third stream and they belong to the brand, not the supplier. A retailer's customer returning a carrier is the brand's commercial problem unless the goods are defective, in which case it becomes a defect return under the warranty. The supplier should not be asked to absorb a consumer's change of mind, and a well-drafted policy says so.

The practical structure is therefore three clauses: a warranty clause covering defect over time, a return clause covering authorised physical returns, and a non-conformance clause covering a shipment that fails at inspection. Most contracts have the first and lack the other two.

A warranty answers "is this a defect"; a return policy answers "what happens to the box"; three clauses, not one.

Return Authorisation and Classification at Intake

Every return starts with an authorisation, and the authorisation is where the classification happens. Without one, goods arrive unlabelled, untraceable and unbudgeted, and the receiving side has to reconstruct what they are — which is the most expensive way to run a returns process.

An RMA number is issued against four fields: the original purchase order or invoice reference, the lot or batch code from the sewn-in label, the quantity by SKU, and the claimed reason with a code. The reason code is the important one, because it routes the return: defect, overstock, wrong item shipped, specification change, transit damage, or consumer return passed through. Six codes cover essentially everything.

Classification at intake is a physical step, not a paperwork one. Goods are counted, photographed before opening, and sorted into resaleable, repairable, scrap and quarantine. The photograph before opening is what settles a dispute about whether damage happened in transit or in a warehouse, and it costs nothing.

Quarantine is the step most processes skip and it exists for one reason: a defect return may indicate a lot problem. Any R1 return should be quarantined until the lot has been checked, because releasing a possibly-affected unit back into stock converts a contained issue into a distributed one.

Timing targets keep the process honest. RMA issued within 24 hours of request, goods despatched within 10 working days of authorisation, intake inspection within 5 working days of receipt, and credit issued within 15 working days of intake sign-off. Those four numbers are the whole service level.

Unauthorised returns should be refused, and the policy should say so plainly. Goods arriving without an RMA are untraceable, they disrupt a warehouse, and accepting them once establishes a precedent that makes the process unmanageable.

Six reason codes, one photograph before opening, and a quarantine step for defects: that is the whole intake process.

Defect Returns: Evidence, Inspection and Disposition

A defect return is decided on evidence, and the evidence standard should be set out in the contract so that neither party is guessing. The standard that works is the same six-item packet used for warranty claims, plus the physical goods where the photographs do not settle it.

The six items: lot code, wide photograph, close-up of the failure, photograph of the whole product, date of first use, and a short description. Where those do not resolve the classification, the physical unit is requested and inspected at 0-40 USD of freight depending on whether it can be consolidated with a forward shipment.

Inspection of a returned unit follows a defined sequence rather than a general look. Count and identify; photograph as received; check the lot code against the production record; examine the failure against the specification; check for signs of misuse or chemical exposure; and classify. Six steps, recorded on a one-page form, and the record is what makes a disputed claim resolvable months later.

Scrap-without-return is the disposition worth negotiating explicitly, because it is usually the cheapest outcome for both sides. A unit that is unrepairable and unsafe has a freight cost both ways that exceeds its value; agreeing that low-value defect returns are credited against photographic evidence and scrapped in the destination market removes 12-40 USD per claim.

Disposal documentation matters where a unit is scrapped in a regulated market. A photograph of the unit rendered unusable — a cut strap or a defaced label — plus a disposal record is the evidence that a credited product did not re-enter the market through a secondary channel.

Root cause closes the loop. Every R1 return should trace to a lot and a failure mode, and the monthly report should show claim rate by SKU and mode against the contract target. Where a lot exceeds three defect returns of the same mode, the retained reference sample is re-examined and remaining stock is held.

Test methods used to judge whether a returned unit deviates from specification follow published practice from ASTM International, and the quality records behind them are maintained under ISO 9001 as published by ISO. Evidence, sequence, disposition: a defect return decided in that order takes days rather than months.

Commercial Returns: Overstock, Resets and Specification Change

Commercial returns are the larger stream by volume in most pet programmes and the one with no warranty behind them. They are a commercial negotiation rather than a quality process, and the terms have to be agreed before the situation arises, because negotiating a restocking percentage on goods already shipped is a weak position for the supplier and an arbitrary one for the buyer.

Overstock is the common case: a buyer ordered 3,000 and sold 1,800, and wants to return the balance. The standard commercial answer is a window of 30-90 days from arrival, goods in resaleable condition with intact retail packaging, a restocking fee of 15-30%, and buyer-paid freight. Below a certain value the return is not worth making at all, which is why most policies set a minimum return value of 200-800 USD.

Retail resets are the second case and they have a longer cycle. A retailer dropping a line will return the shelf stock at the end of a season, typically 90-180 days after delivery, often with the goods opened and shop-soiled. The terms differ: a higher restocking fee of 25-45%, a repack cost, and sometimes a negotiated shared loss rather than a credit.

Specification change is the third case and the one where the supplier usually bears some responsibility. Where a buyer's specification change makes existing stock unsaleable — a branding change, a colour change or a compliance-driven change — the fair allocation is a shared cost: supplier absorbs the material cost, buyer absorbs the freight, and the credit runs 85-100%.

Condition criteria have to be written rather than implied. "Resaleable" should mean: retail packaging present and unmarked, all components and documentation present, no pet hair or odour, no signs of use, and no modification. A unit that fails any one of those is not resaleable, and a dispute about condition is the most common argument in a commercial return.

Pet hair is worth naming specifically, because it is the condition issue unique to this product category and it is absolute: a carrier with pet hair in it cannot be resold as new at any price. Returns accepted without inspection on that criterion produce a 100% loss on those units.

Volume caps protect both sides. A cap of 10-20% of the previous twelve months' purchases on commercial returns in any twelve-month period prevents a return policy from becoming an inventory-financing arrangement, which is what happens when a policy has no cap.

Commercial returns are negotiated before the goods ship: window, condition, fee, freight and an annual volume cap.

Freight, Duty and Customs on Returned Goods

Returned goods are a customs event, and the customs treatment of a return is the part of the process that surprises people. Goods already imported, duty paid and then re-exported do not automatically attract a duty refund; obtaining one requires a specific procedure and, in many markets, a formal application.

The direction of freight is the first allocation decision and it follows the reason code. On a defect return the supplier pays; on a commercial return the buyer pays; on a specification change it is usually split. That is simple to state and worth stating, because an unstated freight allocation on a 2,000-unit return is a five-figure disagreement.

Duty drawback or return relief is the second issue. In the US, unused merchandise exported within a defined period may qualify for a duty refund through a drawback claim; in the EU, returned goods relief may apply where goods are re-exported within three years in the same state. Both require documentation — the original import entry, proof of export and a matching of quantities — and both take weeks.

The practical consequence is that a large return is rarely worth moving physically. A 1,500-unit return at 3.5 kg each is over five tonnes; freight, duty paperwork and handling routinely exceed the credit value on a low-unit-price product, which is why the negotiated outcome is often a credit with goods remaining in the destination market.

Consolidation is the main cost control. Returns held until they can travel with a forward shipment, or consolidated quarterly, cut per-unit freight by 40-70% against despatching each return as it arises. A quarterly return cycle is the standard arrangement in mature programmes.

Documentary controls are what make relief claims succeed. The original commercial invoice and import entry numbers, the export declaration, and a packing list matching the RMA must all reconcile; a mismatch in quantities is the most common reason a drawback or relief claim is rejected.

Where goods are scrapped rather than returned, the same documentation is needed for a different reason: a destruction certificate or photographic evidence is what lets a customs authority accept that exported or destroyed goods were not sold locally.

On a low-unit-price bulky product, the freight and paperwork of a physical return often exceed the credit — negotiate a credit-without-return clause.

Restocking, Credit Notes and Settlement Mechanics

The restocking fee is the number buyers dislike and suppliers need, and it exists for a defensible reason: a returned unit has already incurred handling, and reversing it costs real money. Naming the costs makes the fee negotiable rather than arbitrary.

The costs are: intake inspection at 0.10-0.35 USD, repackaging where packaging is damaged at 0.60-1.80 USD, re-labelling where a barcode or lot label needs replacing at 0.05-0.25 USD, warehouse handling at 0.08-0.30 USD, and the administrative cost of the credit at 5-30 USD per return event. Against a 12 USD unit, a 20% restocking fee is 2.40 USD against a real cost of 0.85-2.70 USD, which is roughly right.

Credit notes are the settlement instrument and they need a defined form: a reference to the RMA, the original invoice, the quantity and SKU, the gross value, the restocking deduction, any freight adjustment, and the net credit. Six fields, and a credit note without the original invoice reference is a reconciliation problem for both parties.

Settlement timing is a working-capital question and should be stated. The common arrangements are a credit against the next order, a credit note offset within 30-60 days, or a refund where there is no open order. Credit against the next order is cheapest for the supplier and usually acceptable to a buyer with a reorder cycle.

Currency and price basis matter where prices have moved. A return credited at the original invoice price protects both parties against a price change; a return credited at current price exposes one of them. The usual position is original invoice price, stated explicitly.

Reconciliation is the administrative step that prevents a build-up of unapplied credits. A quarterly reconciliation of issued credits against applied credits, with anything older than 180 days escalated, keeps the ledger clean and avoids the awkward conversation about a credit neither party can find.

A restocking fee of 15-30% is not arbitrary: it is 0.85-2.70 USD of real handling against a 12 USD unit.

Non-Conformance: What Happens When a Shipment Fails

Non-conformance is the third instrument and the most consequential, because it concerns a whole shipment rather than a few units. It is triggered when a lot fails the agreed inspection standard, and its remedies are different from both warranty and returns.

The trigger should be defined against the agreed inspection: a lot failing final random inspection at AQL 2.5 for major defects, or a buyer's incoming inspection finding a rate above the agreed threshold. Defining the threshold is essential, because a buyer inspecting to a tighter standard than the one agreed will find more defects and call the lot non-conforming.

The remedies are graded by severity. For a minor non-conformance affecting a small share of units, the remedy is a sorting operation at the destination with the cost charged back, or an allowance against the invoice. For a major non-conformance affecting function or safety, the remedy is rework, replacement or rejection with a full credit.

Sorting at destination is usually the right remedy for a moderate non-conformance and it is worth understanding why. Re-shipping a container costs 3,000-5,500 USD in freight plus 35-50 days of production; sorting 3,000 units at destination costs 600-1,800 USD in labour and takes days. The arithmetic favours sorting except where the defect cannot be judged by eye.

Rejection is the remedy of last resort and it should be bounded by a time limit. A right to reject that can be exercised at any time after arrival is a right to hold inventory at the supplier's risk; the standard position is that rejection must be notified within 14-30 days of arrival, with evidence.

Safety escalation sits outside the commercial remedies and it should be immediate. Where a non-conformance indicates a hazard, the obligation runs to the market authority rather than to the contract — in the US through CPSC — and the supplier should be notified the same day rather than at the next review meeting.

Non-conformance is graded, not binary, and sorting at destination is usually cheaper and faster than re-shipping.

Channel-Specific Returns and Programme Notes

Returns behave differently by channel, and a policy written for one channel will not work in another. Three channels account for most pet carrier volume and each has its own mechanics.

Retail chargebacks are the first. A large retailer does not ask for an RMA; it issues a chargeback against a remittance for a defined reason — late delivery, missing label, damaged goods, or a compliance finding — and deducts it. Chargebacks are typically 25-300 USD per incident plus the value of goods, they arrive 30-90 days after the event, and disputing one requires documentation within a defined window. The supplier's practical contribution is preventing them: correct labelling, correct carton quantities and on-time delivery.

Fulfilment-by-Amazon returns are the second, and they are driven by the marketplace's own customer return policy rather than by the brand's. Units come back to a fulfilment centre, are graded sellable or unsellable, and unsellable units accrue storage until a removal order is placed. The cost is a per-unit return processing fee plus storage plus removal, and a brand's return rate above the category threshold affects account standing.

Direct-to-consumer returns are the third and the most generous by necessity, because a consumer return window of 30-60 days is a competitive requirement. The economics are driven by whether a returned unit can be resold: a carrier returned unused with intact packaging is a 100% recovery less 3-8 USD of handling, while one used by a pet is a total loss.

Return rate itself is a design variable, and the largest driver in DTC is sizing. A carrier that arrives and does not fit is returned at the brand's cost; a size chart with three internal dimensions and a pet weight band, plus a fitting diagram in the manual, measurably reduces it.

Reporting closes the loop. A monthly return report by channel, reason code and SKU, with a cost per return, is the management tool that turns a policy into a process. Programmes that measure it usually find one SKU or one reason dominating, and one fix recovers most of the cost.

Our production team supports return programmes with lot-level production records, retained reference samples and intake inspection support through the SGS-verified production base under ISO 9001 and BSCI coverage, with prototypes in 6-10 working days, bulk production 35-50 days after sample approval, final random inspection to AQL 2.5, T/T 30/70 and FOB Xiamen. Three instruments, six reason codes, one monthly report: that is a return policy that works.