Customised Pet Carrier Brand Kit

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

A personalized pet carrier programme needs the artwork file and the variable data list locked before sampling, otherwise the first shipment slips a full cycle.

A Kit Is an Inventory, Not a Mood

The word "kit" invites a folder of inspiration: photography, a palette, a typeface, a paragraph about the brand's values. Those things are useful in a creative conversation and useless in a production one, because nobody can manufacture from them. What manufacturing needs is an inventory of defined assets, each with a format, a tolerance and an owner.

The practical test for whether an asset belongs in the kit is whether two people could produce the same result from it without talking. A vector logo file passes. A photograph of a competitor's bag fails. A Pantone reference with a stated tolerance passes. The phrase "premium and understated" fails. This is not a criticism of the mood board; it is a statement about which document is used for which job, and the mood board belongs in the brief rather than in the kit. The brief side of that distinction is set out in the brief template.

The kit has a second purpose that is easy to overlook: it is the artefact that survives staff turnover. The person who wrote the original brief leaves, a new designer arrives, and the re-order is produced from the kit. If the kit is incomplete, the re-order is produced from a photograph of the first one, and that is how a range drifts a shade per season until somebody notices.

Fourteen assets make up a complete kit for a carrier programme. Six are the identity in its variants. Three are colour. Two are typography and copy. Three are the physical standards that make the first three verifiable. Most brands have eight of the fourteen and do not know which eight.

Logo Files: Three Variants and One Trap

A logo is not one file. It is a set of files prepared for different reproduction processes, and supplying one file for all of them is the single most common defect in an incoming kit.

The three variants that matter are the full colour lockup, the single colour lockup, and the simplified or icon variant. The full colour version is used for print and for the retail carton. The single colour version is what embroidery, weaving, debossing and plating actually consume, because those processes have one colour of thread, yarn or metal — a four-colour logo cannot be embroidered as drawn, and attempting it produces the muddy result everyone has seen. The simplified variant exists because at small sizes a full lockup loses its detail: below about 25 millimetres a wordmark usually becomes unreadable and the symbol has to carry the identity alone.

The trap is the file format. A vector file has mathematical outlines that can be scaled to any size and converted into a stitch path, a weave pattern or a metal die. A raster file — a PNG or a JPEG — is a grid of pixels, and converting it back into outlines is manual tracing work that takes a designer one to three days and introduces errors. A logo supplied as a 72 pixel PNG from a website header is the classic version of this, and it is responsible for more slipped first samples than any other single cause.

One more requirement: outline the fonts. A vector file that references a font the recipient does not have will substitute that font on opening, and the substitution is silent. The logo will look nearly right and will be wrong, and nobody will notice until the embroidery is digitized.

Colour: From a Reference Code to Dyed Fabric

Colour is where a brand kit most often fails, and it fails in a specific and predictable way: a reference that works on paper does not survive translation into a textile.

A Pantone code is a spot ink specification for printing on paper. Ask a dye house to match it on 600-denier polyester and you are asking for a metameric match — the same appearance under one light and a different appearance under another. On a textile, a match within Delta E 2.0 is achievable in a lab dip and is the tolerance we work to. Below Delta E 1.0 the cost rises steeply for no perceptible gain; above Delta E 3.0 the difference is visible on a shelf when two colourways sit next to each other.

Coating changes everything again. The same dye on a fabric with a polyurethane coating reads 5 to 10 percent darker and more saturated, because the coating changes the surface reflectance. A brand kit that specifies a colour without saying whether the surface is coated, uncoated, or heathered is specifying a colour that cannot be matched consistently.

The resolution is the physical standard, and it is the single highest-value item in the colour section. A retained swatch of fabric, or better, a retained sample of an existing product, takes precedence over any code, because it is the actual appearance rather than a description of one. We request a physical standard for every colourway in a programme, we retain it under controlled light, and we refer to it at inspection. Codes are the fallback, not the primary.

Where a colourway has to run across several components — shell, webbing, binding, zip tape, printed label — the kit should state which one is the reference and whether the others are required to match it exactly or to sit in a stated relationship. "Webbing two shades darker than the shell" is a usable instruction. "Matching trim" is not, because zip tape is sold in a standard range and exact matching usually requires a dyed-to-order minimum that a first order cannot absorb.

Typography and the Personalization Constraint

Type in a brand kit usually means the typeface used in the brand's communications, and for a carrier programme that matters in two places: the copy printed on the retail packaging, and any personalization the product carries. The second is the harder one, because it is a system rather than a drawing.

For packaging, the requirement is a licensed font file or a named commercial face. An unlicensed desktop font used in production artwork is a licensing exposure, and it is a real one: foundries do pursue commercial use, and the artwork is already printed by the time it surfaces. Where a brand cannot supply the licensed file, the fallback is to set the copy in a licensed substitute with the same metrics and have it approved — which is a decision, not a workaround, and should be recorded in the kit.

For personalization, the kit needs four rules rather than a typeface. Which face. Which case — and all caps is usually correct below ten millimetres, because lowercase forms spend their height on ascenders and descenders. What the maximum field width is, with the height allowed to float rather than the width. And what happens when a name exceeds the field, which should be a published rule rather than an operator decision. Those four rules are what a variable-data programme runs on, and without them every unit is a small judgement call. The typographic floors for small-format work are set out in personalized embroidery options.

The character set is part of the same field. Accented Latin characters are routine; non-Latin scripts depend on the process; and the kit should state plainly which are supported and what the order form does with an unsupported character. Silence here produces a substitution character in a customer's pet's name, which is the most visible possible failure in this category.

The Hardware Card and the Trim Set

Hardware is the part of the brand kit that most brands have never thought about and the part that most visibly separates a considered product from an assembled one. Every metal component on a bag — zip slider, zip teeth, buckles, D-rings, eyelets, feet, rivets, a branded puller — comes from a supplier, and each supplier has its own version of gunmetal.

Approved individually, each component matches itself and may match nothing else. The control is a hardware card: every metal component mounted together on one card, approved as a set, and referred to at inspection. It costs an afternoon to produce and it removes the single most common visual defect in multi-component soft goods, which is a bag where the plating shades disagree by a fraction that the eye reads as cheapness.

The kit should therefore specify the finish set by name — polished nickel, gunmetal, black nickel, antique brass, matte chrome — and say which components carry which. It should also state whether a branded element is required: a stamped or lasered mark on a zip puller or a buckle is one of the few places where a brand mark can be put into metal rather than onto fabric, and it does not wear off. There is a tooling cost and a minimum, and both should be in the kit rather than discovered at quotation.

Trim is the third element: webbing, binding tape, zip tape, elastic, mesh. Each has a colour, a width and a texture, and each is normally purchased from a standard range rather than custom-dyed. The kit should state the trim colour relationship to the shell — exact match, tonal, or contrast — because an exact match on zip tape usually requires a dyed-to-order minimum of 1,000 to 3,000 metres, which a 500-piece order will not consume.

Packaging Artwork and Retail Assets

Packaging artwork is the asset with the longest lead time in the kit and the one most often started last. It cannot be started last, because it depends on the market list for its mandatory markings and on the carton specification for its dimensions, and both of those are settled well before the bag is finished.

The kit should contain print-ready artwork with bleeds and a defined die line, a stated substrate and finish, and — separately — a list of the mandatory markings by market. That list is the item that gets missed. Importer identification with a postal address for the EU, batch or traceability coding, material composition marking where required, fibre content where a textile declaration is required, and any country-of-origin marking. A carton printed for one market and used for another is not a documentation nicety; in the stricter jurisdictions it is a withdrawal risk. Where a first order is too small for per-market print runs, the usual answer is a single artwork carrying the union of all required markings. See retail-ready packaging for the specification detail and retail box design for the structural side.

The second retail asset is the hang tag and label set, which is where the brand voice usually lives on a pet product. The kit should specify the tag shape, stock, finish and string, plus the copy. The copy field is where claims enter the programme, and every claim on a tag is a test that has to be booked — "water resistant", "tested to 12 kg", "machine washable" each imply evidence. Listing them in the kit means they are costed and scheduled rather than discovered on the artwork.

The third is the insert card, which on a personalization programme does real work: it explains how to add or change a name, and it converts a one-off purchase into a second transaction. It is the cheapest asset in the kit and often the highest returning.

Photography and Content Direction

Photography is not a production asset and it belongs in the kit anyway, for one reason: it has to be taken against the pre-production sample, which exists four to six weeks before the goods land, and if the direction is not written down by then the window closes and the listing goes live three weeks late.

The kit does not need a shot list as detailed as an agency brief. It needs six things: the angle convention per SKU, so a range looks like a range; the lighting mood, stated as a reference rather than as an adjective; whether the product is shown empty, with an animal, or with a handler; the scale reference convention, which on a pet bag is what actually communicates size to a customer; the background treatment; and the crop and aspect ratio per channel. Six decisions, made once, applied to the whole range.

The angle convention is the one that does the most work. A range photographed from four different angles does not read as a range, and it underperforms on a category page where the eye is comparing. Pick one angle per product type and hold it across every SKU and every season, including reorders.

The animal question has a practical dimension. If the product is photographed with a dog, the dog's size has to be stated relative to the size class, because a photograph of a medium dog in a large bag is a sizing complaint waiting to happen. Where the programme carries fit data, the photography should be consistent with the published size chart, and the fitting guide is the source for that.

Version Control: The Failure Nobody Budgets For

Every asset in the kit will change, and the change is where programmes come apart. A designer updates the logo, a new file reaches the embroidery digitizer, an older file is still with the label weaver, and three months later the woven label and the embroidered logo disagree. Nobody made an error; two people used two versions.

The control is a version log: a dated list of every change, what changed and who approved it, held with the kit rather than in someone's inbox. Files are named with a version and a date, and the previous version is archived rather than overwritten. That is the whole mechanism, and it is the least glamorous and most load-bearing part of the kit.

The second half of the control is an explicit statement that the approved physical sample outranks the digital file. Once a sample is signed off, it becomes the reference for colour, placement and scale, and a later revision to the artwork is a change against that sample rather than a correction to it. This prevents the situation where a brand quietly updates its logo and expects the current production run to follow.

What we return at the end of a programme is a counterpart kit: the approved hardware card, the retained colour standard, the specification with the version log, the test record, and the approved sample. That pack is what a reorder runs from, and it is why a reorder carries almost no setup cost — the assets already exist and are already agreed. The commercial consequence of that is set out in the unit economics note, and the point in the calendar where the kit has to be complete is shown in the launch sequence.

What to do if the kit is incomplete

Most kits arrive at eight of the fourteen assets, and that is workable if the gaps are known. Send what you have, mark the six that are missing, and the gaps go onto the open-items page with an owner and a date. What does not work is sending a folder of mixed-format files and assuming the recipient will work out which is current. A brand that has no single-colour logo variant should say so rather than sending the full-colour one and hoping; the single-colour variant is prepared in a day and is the one that four of the five decoration processes actually consume.