A first sample is the most expensive way to discover that a supplier cannot make the product. It costs a pattern, a set of materials, a courier and several weeks, and it produces nothing that can be sold. Checking capability before the sample costs a few questions and a document review, and it is the single highest-return hour in a first production programme.
Clothing Manufacturer for Startups: what to confirm before you send a tech pack
Confirm the category, the machinery for the required operations, the minimum for the fabric and the construction, and the site that would produce the order. Those four decide whether a sample is worth commissioning.
The check is deliberately narrow. Category, because a factory outside its family will develop your product from scratch and charge for the learning. Machinery, because each construction method needs specific equipment. Minimum, because a programme that cannot reach it will not be made regardless of how well the sample turns out. And site, because an order cannot be assessed at a factory nobody will name.
None of those four requires a visit or an audit. Each is answered in a first exchange by a supplier that knows its own floor, and the answers are specific enough to compare between candidates. Where any one of them is vague, the sample is a test of the supplier rather than of the product, and the cost is the same.
Documentation a knit apparel order should arrive with, and who prepares it
A first order arrives with a document set, and knowing which party prepares each item prevents the gap that appears at the packing table. The buyer prepares the specification, the measurement table with tolerances, the artwork and the label text. The factory prepares the sample record, the measurement readings, the inspection report and the packing list. The mill or the fabric supplier prepares the fabric specification, the lot numbers and any test report.
Two documents are the ones a startup most often lacks. The measurement table with tolerances, because a chart without them cannot be checked and turns every inspection into an opinion. And the approved sample record, because an approval without a retained reference cannot be enforced later.
The material evidence sits alongside those. Where a claim will appear on a label, the report has to name the material in the garment rather than a comparable product, and OEKO-TEX Standard 100 states the article classes its certificate covers. The vocabulary for the physical properties being tested is published in the ASTM standards programme for textiles, and a laboratory publishes the scope it applies to apparel, as Intertek’s textiles and apparel testing range illustrates.
Where a first order is placed after any sustainability or workplace claim, the supporting documents should be requested with the order rather than after it, since the frameworks that publish their scope make verification routine: Sedex holds SMETA reports for workplace assessment, and the procedures customs authorities are expected to maintain for a cross-border consignment are framed by the WTO Trade Facilitation Agreement.
| Document | Prepared by | What it protects |
|---|---|---|
| Specification and measurement table | Buyer | The product being ordered |
| Fabric specification and lot numbers | Mill or fabric supplier | Material consistency and traceability |
| Approved sample and its record | Both parties | The standard later deliveries are compared with |
| Inspection report | Factory or third party | The decision to ship |
| Packing list and invoice | Factory | Quantity, value and the freight booking |
How colour approval works on knit apparel, and where it usually slips
Colour on knitwear is approved in stages rather than at once. A colour reference is issued, the mill produces a lab dip or a strike-off, the buyer approves it against the reference under a defined light source, and only then is the fabric dyed. The approval is a decision with a date, which is why a slow approval moves the dyeing and therefore the whole schedule.
Three points are where the process slips. Approving a lab dip from a photograph, because a photograph is not a colour reference and the dyeing will match the dip rather than the image. Approving under an undefined light source, because a shade that matches in one light can read differently in another, and the reference conditions should be agreed rather than assumed. And approving a fabric swatch and then specifying a different fabric for production, which invalidates the approval entirely.
One further point is specific to knits and worth knowing before the first order. Because a knitted fabric is dyed as a batch, two batches of the same approved colour can differ within the shade band, and a reorder placed months later will come from a new batch. Where a programme expects reorders, asking whether the mill can hold a standard and re-match to it is a question worth asking before the first order rather than after the second.
Trims and findings on knit apparel: what to specify and what to leave to the factory
Trims divide into two groups, and the division decides who specifies them. The components that touch the customer or carry a claim should be specified by the buyer: the main and care labels with their wording, the hangtag, the thread where colour matters, and any elastic or ribbing whose recovery affects fit. Those affect the product and often the compliance position.
The components that are genuinely interchangeable can be left to the factory, provided the performance requirement is stated rather than the part number. A drawcord end, a sewing thread for a concealed seam or a plain polybag can be selected by the factory if the buyer has said what it has to do. The distinction is between specifying a requirement and specifying a supplier.
One practice keeps the list under control. Name the performance requirement against each item, such as the recovery expected from an elastic, the load a zip has to carry or the temperature a heat-applied label must tolerate. Where the requirement is stated, a substitution that meets it is usually acceptable and a substitution that does not is detectable; where only a part reference is given, any substitution is a change and the conversation becomes commercial.
Planning a second knit apparel order: what to reuse and what to re-approve
The second order is where a programme becomes efficient or does not, and it turns on a short list of items that carry forward. The pattern, the grade and the marker carry forward if their ownership was settled. The measurement table and the tolerances carry forward while the construction is unchanged. The label artwork carries forward if the market requirement has not changed. And the specification carries forward if it was versioned, because the current version is identifiable.
Four items have to be re-approved rather than carried. The fabric, because a new batch is a new lot even where the specification is unchanged, and the shade approval applies to a batch rather than to a colour. The sample, where any construction or trim change has occurred, because a change reopens the question the approval closed. Any certification that has expired while the programme was running. And the label text, where a market requirement has been revised in the interim.
The efficient version of a second order is therefore a short confirmation rather than a new development, and the document that makes it possible is a versioned specification plus a retained approved sample. The protocol for running an approval so that it holds across orders is set out in this sample approval protocol for wholesale orders.
How knitwear minimums behave: per style, per colour and per size
A minimum for knitwear is expressed at three levels, and they behave differently. Per style, which is usually the smallest threshold and is set by the line arrangement. Per colour, which is set by the dyeing batch and is normally the binding constraint on a first order. And per size, which is rarely a minimum in its own right but becomes one when a size ratio cannot be met within the pack configuration.
The practical consequence is that adding a colour is expensive and adding a size is usually not, which inverts the assumption many first-time buyers make when planning a range. A palette should be chosen against the colour minimum rather than against a merchandising idea, and where a programme needs several colours at low volume, the two available routes are to consolidate them or to accept a higher unit price at a smaller batch.
One structural option is worth knowing about before the first order rather than after. Where a colour is expected to repeat, a mill may be able to hold a standard and re-match against it, which is what makes a seasonal reorder possible without re-opening the colour. The way the thresholds work, and how they are expressed in a quotation, is set out in this explanation of factory minimum order quantity.
How to find a manufacturer to start a clothing brand?
Filter on category, then verify the four facts before sampling.
Start with factories that already run your category, verify their machinery, minimum and production site in the first exchange, and only then commission a sample.
The filter matters because the search itself is not the hard part; directories and exhibitions produce hundreds of names. What reduces the list is a category filter applied before contact, since a factory running your product family every week will price the labour accurately and develop less. After that, the four verification facts, category, machinery, minimum and site, separate the suppliers worth sampling from the rest.
Two structural questions are worth asking at the same time. Which processes are on site and which are subcontracted, because a subcontracted step is a schedule the factory does not control. And who owns the pattern for a style developed with them, because for a startup the pattern is the asset that makes a second order cheap. Where the pattern belongs to the factory, the style does not travel. The questions that establish what a supplier can commit to are described in this guide to the questions to ask a clothing manufacturer.
What drives the cost of a first production run?
Fixed costs spread over few units, not a higher unit price.
A first run costs more per unit because sampling, pattern work, label and packaging setup and small fabric batches are spread across a small quantity, and none of those costs fall when the run is short.
Three fixed costs dominate a startup order. Development, which includes the pattern, the grade and the sample rounds, and which is either invoiced separately or amortised into the unit price. Setup, which includes label artwork, print or embroidery screens and packaging tooling, and which is charged per version rather than per unit. And material, because a small fabric batch is priced at a minimum rather than at a rate, so the effective price per metre is higher than it would be at volume.
The commercial consequence is that the first order should be planned to reach the thresholds it can reach, and that the second order should reuse everything possible. Reaching a tier by buying inventory that will not sell is not a saving, but consolidating a palette to meet a colour minimum usually is, because the fixed cost is removed rather than shifted. Where the price is quoted without naming its delivery term, the comparison is also incomplete, since the ICC Incoterms rules define what each term includes and therefore what the number covers.
Conclusion
Capability can be verified before a sample is commissioned, and the check is short: category, machinery, minimum and site. Beyond that, a first programme turns on a small number of documents and a small number of decisions. Specify the trims that touch the customer and the compliance file, leave the interchangeable components to the factory with a performance requirement attached, approve colour against a reference under an agreed light, and version the specification so that the second order is a confirmation rather than a development. The first order costs what it costs because of fixed costs, and the way to make the second one cheaper is to have decided, in writing, what carries forward.
FAQ
What should I check before commissioning a sample?
Whether the factory already makes your category, whether the machinery for the required operations is present, what the minimum is per style and per colour, and which site would produce the order. Those four decide whether the sample is worth paying for.
Why do first orders cost more per unit?
Because fixed costs are spread across a small quantity: pattern and grade work, sample rounds, label and packaging setup, and small fabric batches priced at a minimum rather than a rate. None of those costs fall when the run is short.
What has to be re-approved on a second order?
The fabric, because a new dye batch is a new lot even at the same specification. The sample, where a construction or trim has changed. Any expired certification. And the label text, where a market requirement has been revised.
Should the factory or the buyer specify the trims?
The buyer should specify anything that touches the customer or carries a claim, such as labels, hangtags and elastics that affect fit. Interchangeable components can be left to the factory provided a performance requirement is stated rather than a part number.
Next step. If you are sourcing for a first range, send the category and where the product is now. The team can confirm which operations the construction needs, what the minimum would be per colour, and what is worth verifying before a sample is cut.
