Private Label Pet Bags: Launch Sequence

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

A custom pet bag programme is usually a size ladder rather than a single style, and the ladder is what determines how many colourways you can afford.

Phase Zero: The Commercial Case Before the Brief

Most launch delays are decided before anybody writes a brief, in a conversation that was never had. Somebody in marketing wants a bag, somebody in buying has a target cost, and nobody has written down what the range has to earn. Phase zero is where that gets resolved, and it costs three days.

The document is one page with five numbers on it: target retail price, target landed cost, target margin, first-season volume and the date the product has to be on shelf. Those five constrain everything downstream, and the fifth one is the one that does the work. A launch date set backwards from a retail line review is hard, and it determines whether you can afford a custom tooling route at all, because a custom mould or die adds 8 to 14 days that cannot be compressed by anyone.

The second thing phase zero resolves is the range shape: how many sizes, how many colourways, and whether the first drop is one hero product or a family. A family launch multiplies every downstream step by the number of variants — four sizes in three colourways is twelve specification sheets, twelve sample rounds and twelve compliance entries, not one. Brands routinely underestimate this by a factor of three, and it is the most reliable predictor of a slipped launch that exists.

Phase zero also resolves one question that otherwise resurfaces at every gate: who can say yes. A programme needs two named signatories — one for design, one for technical — and a named deputy for each, because the largest single cause of slip is not a manufacturing delay but an approved sample sitting on a desk while the only person entitled to sign it is at a trade show. Naming the deputy costs nothing in week one and recovers an average of eleven days later.

The output of phase zero is a signed one-page commercial case. It is not the brief. It is the thing the brief has to satisfy, and when a decision is contested in week eight — and one always is — it is the document you go back to. Our brief template is structured to be filled in directly against it.

Phase One: Concept Sampling and What It Is Actually For

A concept sample is not a product. It is an argument, made in three dimensions, about whether the idea is buildable at the cost you need. Treating it as a product — photographing it, showing it to a retailer, putting it in a line review — is the single most common way to lose six weeks.

What a concept sample legitimately answers: does the silhouette work, does the panel layout make sense, does the material behave the way the mood board assumed, and does the target cost survive contact with a real pattern. What it cannot answer: whether the product will pass a drop test, whether the colour matches the Pantone reference, or whether the closure will survive a season. Those are specification questions and they belong in phase three.

Concept samples are produced in 6-10 working days and we recommend two rounds. The first round is deliberately loose — one prototype, in a stock colour close to the intent, built to explore geometry. The second is tighter: the agreed geometry in the intended material, with the hardware the programme is likely to use. Two rounds cost about two and a half weeks and they remove almost all of the geometric surprises that would otherwise surface in pre-production, where a change costs five times more.

The gate out of phase one is a written geometry approval signed by whoever owns the design, with a photograph annotated by hand. Not an email saying "looks good". The annotated photograph is what the pattern maker works from, and an email is not a specification. If your organisation cannot produce a signed geometry approval in under 48 hours, put that in the plan as a known risk, because it will happen again at every subsequent gate.

Phase Two: Specification Lock

This is the phase that decides whether the launch lands, and it is the phase that gets skipped. A specification is a document that describes the product completely enough that two different people could build it and get the same result. It is not a photograph, a sample or a mood board, and none of those three is a substitute.

A carrier specification runs to roughly eleven pages and covers: every panel with its material reference and coating; every dimension with its tolerance; every hardware item with finish and pull-force; every seam with stitch density; every print or embroidery with its method, colour reference and placement; the packing and carton specification; and the test standard each claim is made against. Eleven pages is not excessive. A bag has more specification surface than most people expect, because it is a soft product with a structural claim, and every dimension can vary.

The critical discipline in phase two is that nothing is "TBC". An open item in the specification becomes a change order in production, and a change order in production costs between four and eleven days depending on where it lands. We log every open item on a single page with an owner and a date, and we do not cut material until the page is empty. This is the gate: an empty open-items page, not a nearly empty one.

The other discipline is to test at specification stage rather than after. If the product will carry a claim — water resistant, airline compliant, tested to a load — the test is booked against the specification before bulk, because a test failure discovered after 500 units are cut is not a test failure, it is a write-off. Where a claim touches a regulated area, the evidence is gathered against US Consumer Product Safety Commission requirements at this stage rather than at the end.

Phase Three: Pre-Production Sample and the Approval Gate

The pre-production sample is built on the production pattern, from production materials, on the line that will run the order. That is the entire difference between it and a concept sample, and it is why this sample is the one you approve and the one you photograph.

Three things get verified against it. The first is dimensional: the finished unit is measured against the specification, panel by panel, and any dimension outside tolerance is corrected on the pattern rather than accepted. The second is aesthetic, and it is where colour lives: the approved sample is the physical colour standard for the whole programme, and it is retained under controlled light. A Pantone reference alone is not sufficient for a textile, because the same reference dyed on polyester and on nylon is two different colours, and dyed on a coated shell is a third.

The third is functional, and it is where the closure, the strap attachment and the floor are loaded to the rated specification. A strap root that pulls at 1.5 times rated load on one sample will pull at some fraction of that across 500 units, because stitch tension varies; the sample is not the proof, it is the check that the construction is in the right order of magnitude. The real proof is the inspection at the end.

The gate is written approval with the sample physically retained. Once it is signed, the pattern is locked and the material is cut. Everything after this point is either production or a change order, and a change order here costs days rather than hours. This is also the point at which the commercial terms crystallise: 500 pieces per colourway, and bulk production in 35-50 days from the date of that signature.

What belongs in the approval pack

An approval pack is the sample plus five documents: the dimensional report, the colour standard card, the hardware card showing every metal component mounted together, the test record for any claim, and the packing specification. Retailers increasingly ask for this pack, and having it assembled at approval rather than at shipment is the difference between a range that onboards in a week and one that onboards in a month.

Phase Four: Bulk, Inspection and the AQL Conversation

Bulk production is 35-50 days, and the band is wide because the variables are real. Cutting, sewing, assembly and packing are stable and predictable. What moves the date is material lead time — a custom-dyed shell is two weeks longer than a stock colour — hardware lead time, and any second operation such as printing or embroidery that has to be sequenced after assembly.

Inspection happens before release and it is statistical, not visual. AQL 2.5 means the shipment is sampled to a defined plan and accepted or rejected on a defect count, with critical, major and minor defects counted separately. A critical defect — a structural failure, a broken closure, a sharp edge inside the animal's space — is a zero-tolerance attribute and rejects the lot on its own. Major defects are capped at a count derived from the lot size; minor defects at a higher count. What surprises brands is that AQL 2.5 does not mean 2.5 percent defective product is acceptable. It describes the probability that a lot of a given quality passes, and a well-run line passes comfortably.

The second thing that surprises brands is that inspection is the wrong place to fix quality. By the time a carton is packed, the only available remedies are rework and rejection, and both are expensive. The controls that matter are the first-piece check at cutting, the in-line check at the sewing station, and the hardware card at assembly. Inspection is the safety net, not the system.

Where a programme is shipping directly to a fulfilment centre rather than to a brand warehouse, add two days for carton labelling and pallet configuration, and agree the labelling specification at phase two rather than at phase four. Mislabelled cartons are the most common cause of a rejection at a retailer's goods-in desk, and they are entirely a documentation failure.

Phase Five: Freight, Customs and the Landing Plan

Freight is booked at the start of bulk, not at the end, and that single habit is worth about ten days on a sea shipment and about two on an air one. The reason is that the constraint is rarely the vessel; it is the document set that has to accompany it.

Three dates have to line up: the production completion date, the document availability date and the carrier cut-off. They are produced by three different parties and they are almost never aligned by default. A commercial invoice and packing list can be produced in a day. A certificate of origin, a material declaration or a test report can take a week to two weeks, and if any of them is issued against a material that changed during production, it has to be reissued.

Sea freight from Xiamen to a northern European port is typically 28 to 38 days door-to-door including customs; to a US west coast port it is 18 to 25. Air freight is 5 to 8 days and costs four to six times more per cubic metre, which on a bulky, low-density product like a pet bag usually destroys the margin — with one exception. If a launch date is genuinely at risk, air-freighting a partial quantity to open the channel and sea-freighting the balance is often the right answer, because the margin lost on 150 units is less than the margin lost on a missed line review.

The final decision in phase five is Incoterms. On FOB Xiamen terms, the brand controls the carrier and the risk transfers at the rail; we control export clearance and loading. Brands new to importing consistently underestimate the work in that second half, and it is worth pricing a freight forwarder's fee before choosing an Incoterm rather than after.

Phase Six: Channel Assets While the Cartons Are Moving

The longest unproductive gap in most launches sits between the container leaving and the listing going live, and it exists because the content was scheduled after the goods arrived. It can be eliminated entirely, because nothing in the content chain actually needs the finished carton.

Product photography needs a pre-production sample, which exists in phase three — four to six weeks before bulk finishes. Copy needs the specification, which exists in phase two. Size charts and fit guidance need the fit data, which is produced alongside the sample. Lifestyle photography needs a dog and a location, neither of which is on the container. Every asset can be complete before the goods land, and when it is, the listing goes live the day the cartons are received rather than three weeks later.

The exception is packaging artwork, which depends on the final carton and print specification, and compliance markings, which depend on the final market list. Both should be finished at phase three against the approved specification, with the printer booked in parallel with bulk. Where a brand is selling into more than one market, the artwork should carry the union of all required markings rather than a market-specific version, unless volume justifies separate print runs — and at 500 pieces per colourway, it usually does not.

One more asset belongs here and is almost always forgotten: the returns script. If a customer asks whether the bag fits their dog, the answer has to exist before the first order, and it has to be the same answer the size chart gives. Building that script from the fit data is a day of work and it removes the largest single source of early negative reviews on a new range. See retail-ready packaging requirements for the artwork and marking side of the same window.

Where Launch Sequences Actually Slip

Having run this sequence many times, the slips are remarkably consistent, and they are almost never in production. In rough order of frequency, this is where the calendar goes.

The first and largest is internal approval latency on the brand side: the sample sits on a desk for eleven days because the person who can sign it is travelling. This is not a manufacturing problem and no amount of production planning fixes it. The only mitigation that works is to name a deputy signatory at phase zero and to give the deputy the authority to sign.

The second is artwork arriving in the wrong format. A logo supplied as a 72 dpi PNG cannot be printed, embroidered or woven, and converting it takes a designer's time that was not in the plan. The fix is a brand kit issued at phase zero with vector files, and our companion note on the customised brand kit sets out exactly what that pack should contain.

The third is a material change after testing. A shell is substituted for a cheaper equivalent, the test report no longer covers the product, and the document has to be reissued — which in a regulated market can add three weeks. The rule is simple: change the material before you test, never after.

The fourth is scope creep in phase two, where a feature is added after the pattern is drafted. Adding a pocket sounds trivial and it is not, because it changes the panel layout, the stitch sequence and the packing volume. Every addition in phase two should be priced in days as well as in money, and the days should be added to the plan in writing.

The fifth, and the only one on the production side, is the split colourway. A programme of 500 units in one colourway runs cleanly; the same 500 split across four colourways is four material setups, four cutting runs and a longer line changeover, and it adds 5 to 8 days. That is a legitimate commercial choice and it should be a conscious one, made in MOQ planning rather than discovered in week six.