How clothing manufacturing works, from design to finished garment

This is a walk through the process as a sequence of decisions rather than a list of rooms. For each stage: what is being decided, who owns that decision, what it locks in, and what it costs to change your mind afterwards. If you understand only one thing about manufacturing, make it this — the price of a change rises steeply and predictably as the garment moves forward, and it never falls back.

It is written for brands placing work with an overseas manufacturer, so where responsibility sits on the other side of the table gets named explicitly. If you want the shorter version of who does what, what a womenswear manufacturer actually does covers that ground, and our production process shows the same sequence as we run it.

Quick answer: the eight stages and what each one decides

Production runs through eight stages. Each answers one question, and each one narrows what can still be changed afterwards.

Read down that right-hand column and the shape of the whole thing becomes obvious: each stage converts flexibility into certainty. Certainty is what you are buying — but it is also why asking for a change late is expensive rather than merely inconvenient.

Stage 1: the brief, and the two things it must settle

A brief is not a design. A design says what the garment should look like; a brief says what the manufacturer is allowed to decide. Most slow first orders are not the result of a difficult design — they are the result of a brief that left the important gaps unnamed, so both sides assumed the other would fill them.

The gaps that cause the most rework

• Who chooses the fabric. A supplier who is asked to “use a similar satin” will select on availability and price. If drape or weight matters, that has to be in the brief, not discovered at the sample.
• What the target price is. A price target changes construction choices — lining, closure type, finishing method. Withheld, it produces a beautiful sample you cannot afford to run.
• Which sizes are real. A full size range is a different cost structure from three sizes, because grading and marker planning both scale with it.
• What “the same as last time” means. On a repeat, this phrase is doing more work than any other three words in the trade. Name the reference garment and the version date.

Who owns which decision

Notice that the brand's decisions are concentrated early and the factory's later. That asymmetry is not a convention — it reflects which party can still act cheaply at that point. A brief that respects it moves quickly. One that pushes brand decisions to the end pays for them.

Stage 2: pattern and development

The pattern is the first place where the design becomes a set of numbers, and it is where most of the technical risk in the whole project is created or avoided.

Why the block matters more than the pattern

A block is a factory's proven base shape — a fitted shoulder, armhole and torso that has already been corrected on real bodies. A new pattern drafted from an existing block inherits that correctness. One drafted from scratch inherits nothing and has to earn it through fitting rounds. This is the single largest source of difference between two factories quoting on the same style, and it is why a factory with the right block can be faster on a new style than a cheaper factory starting from zero.

Drafted, draped, or both

• Drafted patterns suit clean, structured garments where the measurements carry the shape — a fitted sheath, a tailored bodice, a pencil skirt.
• Draped patterns are built on a stand in cloth, and suit garments where the fabric is doing the work — cowls, asymmetric wraps, soft bias cuts.
• Both is the honest answer for most eveningwear: the bodice drafted so it holds, the drape explored on the stand so it falls.

For a fuller treatment of this stage on its own, how pattern making works in womenswear goes deeper into blocks, grading and pattern retention.

Stage 3: fabric and trim booking

Fabric is where a manufacturing project touches the outside world, and therefore where it stops being fully under anyone's control. A pattern can be redrafted on Tuesday; a cloth that is out of stock is out of stock.

Fabric availability sets the date, not the sewing line

Ask what actually constrains a delivery window and the answer is rarely machine hours. It is whether the base cloth is booked, whether the dye lot is confirmed, and whether the trims exist in the specified finish. A factory with an empty line and no fabric is not fast — it is idle, and the date slips by exactly the length of the fabric lead time, which is the one number in the whole chain that nobody can compress by working harder.

From swatch to bulk: the shade problem

An approved swatch is a promise about a small piece of cloth. Bulk is a promise about hundreds of metres, dyed in one bath, and the two are not automatically the same colour. This is not dishonesty — it is dyeing. The practical defence is to approve the bulk lot, not only the swatch, and to keep the approved swatch in a drawer with a date on it so that a reorder two seasons later can be measured against something physical rather than a photograph on a phone.

Stage 4: sampling

Sampling exists to fail cheaply. A sample that reveals a problem on a table costs a week; the same problem discovered in a finished production run costs the run.

What a sample proves, and what it does not

• Proves: that the silhouette works on a body, that the construction sequence is achievable, that the fabric behaves as expected on the stand.
• Proves: that the measurements are achievable in the chosen cloth, with the chosen lining and interlining.
• Does not prove: that the garment will look the same at the twentieth unit. That is a line-consistency question, and it is answered by process, not by a sample.
• Does not prove: that the colour will match a reorder. Only a retained swatch does that.

How to run a comment round so it takes one round

The cost of a sample round is not the sample. It is the days between the sample arriving and the comments leaving. A brand that opens the parcel on Friday and replies on Wednesday has added five days to the calendar while believing the factory is late. Comments sent as five separate messages across four days cost more than the comments themselves.

For the mechanics of multiple rounds — how many are normal, what each one is for — see how sample rounds work.

Manufacturing perspective: what the floor sees and the brief does not

This section is the part a design-side brief tends to miss, and it is where experience actually shows. Three things look fine on paper and cause trouble on a line.

Three things invisible until production

• A seam that cannot be pressed flat. Bulk and layered seam allowance in a curved seam produce a ridge that no amount of finishing removes. It has to be resolved in construction, before cutting.
• A trim that arrives in the wrong finish. The sample uses the one gold zip a supplier had; the bulk order needs four hundred of them. Finish, length and pull type must be specified, not implied.
• A fabric that shrinks unequally. A cloth that relaxes more in the weft than the warp will change the fit of the same size across the run. It is measurable before cutting, and it is expensive to discover afterwards.

Notice the pattern: all three are properties of materials and construction, not of design. That is why a factory's opinion is worth taking before the pattern is finalised, and why a brand that treats the factory as a pair of hands pays for that choice later in the calendar.

The practical version of this is a conversation, not a document. Before the fit sample is cut, a production team that has run the fabric before will usually be able to name the two or three places this particular garment is likely to fight back. Ask for that list. It is not a criticism of the design; it is the cheapest technical review you will ever receive.

Stage 5: cutting

Cutting is the first stage where the garment becomes physical and irreversible. Up to here, everything was a document.

The marker is a cost decision

A marker is the arrangement of pattern pieces on the cloth. How tightly they nest determines how many metres of fabric each garment consumes, and fabric is usually the largest single line in the cost of a dress. Two factories can cut the same pattern with a measurable difference in consumption purely on marker efficiency and the willingness to spend time on it.

That second row is the one worth internalising. Marker efficiency is a real saving right up to the point where it is achieved by cheating the grain line, at which point it becomes a twisting seam that a customer notices in the mirror and returns. A factory that refuses to do it is protecting you from a cheaper, worse garment.

Why cutting errors are the most expensive kind

A sewing error affects one unit and can be undone. A cutting error affects every panel cut from that lay — often every unit of that size in the run — and cannot be undone at all, because the cloth is now smaller than it was. This is the argument for spending time on the cutting ticket and the size ratio before the knife moves, even when that costs a day.

Stage 6: sewing

Sewing is the stage buyers imagine when they think about manufacturing, and the stage where the least irreversible damage happens. It is also where the difference between an organised line and a busy room shows up most clearly.

Line balancing, in plain terms

A garment moves through a sequence of operations. If one station takes twice as long as any other, the whole line moves at the speed of that station and every other operator waits. Balancing means splitting and recombining operations so no single seat is the bottleneck. It is why a factory that makes a hundred styles a month can absorb a new style without a shock, while one that repeats three styles will struggle: balancing is a skill, and it is exercised or it is not.

The specialist seats that decide quality

• The setter. Sleeves, collars and bodice joins. Sets the shape of the garment more than any other seat.
• The invisible zip operator. A seat that either exists at a quality level or does not; there is no middle.
• The beading and embellishment bench. Sequins and hand work, where density consistency is the whole job.
• The presser. Finishing can make an average garment look expensive and a good garment look cheap.

When a brand asks what a factory is good at, these seats are the honest answer. A factory with a strong beading bench can take on an embellished gown that a generalist would struggle to keep consistent, regardless of how the price lists compare.

Stage 7: finishing, pressing and trim

Finishing is where a garment stops being construction and becomes a product. It is also the stage most often squeezed when a schedule is tight, which is a mistake with a visible cost.

• Trimming and thread-cleaning — the loose threads that make an otherwise correct garment read as cheap.
• Pressing and steaming — the step that reveals whether seams sit flat. A seam that only looks flat under a press will not stay flat through shipping.
• Measurements re-checked cold — pressing relaxes cloth, and a garment measured hot is a garment measured wrong.
• Hardware and fastening checks — pull tests on closures, zip runs, hook positions.
• Labelling — care, content and origin, placed where the brand specified, in the language the destination market requires.

Stage 8: inspection

Inspection is not an opinion about quality. It is a defined standard applied to a sample and reported as a number, which is the only version of it that can be argued with productively.

Inline and final answer different questions

The reason both exist is that they have different economics. Inline catches a fault when a hundred units have it and none are finished; final catches the faults that survived. Relying on final alone means discovering a systematic problem at the point where it is most expensive to correct. Inline versus final inspection covers the comparison, and our quality control service describes how the two are scheduled here.

Acceptable quality level, briefly

Most factories inspect against a sampling standard — a defined number of units drawn from the batch, with a defined number of defects allowed before the batch is rejected. The value of the standard is not that it guarantees perfection. It is that it is agreed in advance, so a disagreement at the end is a conversation about measurements rather than about impressions.

Stage 9: packing and documentation

Packing is where a production decision becomes a logistics cost, and where brands most often discover that a garment costed at the factory gate is not the garment they are selling.

• Fold and polybag method — determined by whether the goods are going to a warehouse, a retail floor or a marketplace fulfilment centre.
• Carton dimensions and weight — volumetric weight can be charged instead of actual weight, so a carton that is slightly too large is a permanent surcharge.
• Carton marking — style, colour, size ratio and destination on the outside, because someone has to find this box again.
• Packing list accuracy — the document that customs and your receiving team both rely on.

The commercial point is that packing decisions are cheap before they are made and expensive after. Export packing for dresses covers the detail, and our packaging service sets out what can be specified.

Practical example: the same dress, two different projects

Two brands approach the same factory with the same reference image for a lined satin midi. The garments look identical on paper. The projects behave nothing alike.

The two projects are not separated by budget or by design ambition. They are separated by how much uncertainty was removed before the work started, and by when. Brand B can retrieve the same result — but it will spend the extra time on development and booking, and no factory can compress those two things by working faster in the sewing room.

Common mistakes, by stage

• Brief: sending a mood board and a price expectation in separate conversations three weeks apart.
• Development: treating the pattern as a drawing rather than an asset that will be re-used on every reorder.
• Fabric: approving a swatch in daylight on a phone screen, then discovering the bulk in a fitting room.
• Sampling: approving a sample for silhouette and assuming it also locks the measurements.
• Cutting: pressing for a shorter delivery without confirming the size ratio, then accepting a marker that distorts grain to save cloth.
• Sewing: changing one operation mid-run because it is easier, without asking what it does to the other nine.
• Inspection: inspecting against a feeling rather than an agreed standard, so the report cannot be acted on.
• Packing: specifying the carton at the point of shipping rather than at the point of costing.

How we run this process

Luxudress has manufactured womenswear in Guangzhou since 1995, with pattern, cutting, sewing, finishing and packing under one roof. That structure is not a marketing detail — it is the reason the sequence above can be run as a sequence, with each stage's output becoming the next stage's input rather than a document that has to survive a handover.

1. We read the brief and name the gaps. Before quoting, you get the questions that would otherwise become rework.
2. We tell you where the risk is. If the fabric, the silhouette or the size range is going to make this slow, you hear it at the start.
3. We develop from the block that fits your silhouette. Not from zero, if we already hold one.
4. We sample, and we tell you what the sample proves. Silhouette and construction, clearly separated from anything it does not settle.
5. We confirm the standard before bulk. Measurement, colour and packing tolerance agreed, then frozen.
6. We inspect inline and at final. Faults are caught where correcting them is still cheap.

If you are planning a first run or replacing a supplier who has drifted, send us the design, the target quantity and the delivery window you are working to. We will tell you what is achievable, what is not, and which of the two is worth arguing about. Our production process page shows the same stages with the points where brands usually need to make a decision, and product development covers the front end of the sequence in more detail.

Frequently asked questions

How long does the whole process take?

It depends almost entirely on how much is being developed rather than repeated. In our production, sampling runs around seven to ten days from tech pack approval and bulk production roughly four to six weeks after sample approval. Development before sampling is the variable: a style built on a block we already hold moves quickly, a first style in a new silhouette does not. How long manufacture takes breaks the calendar down stage by stage.

Can we skip sampling and go straight to bulk?

Technically yes, and it is a bet rather than a saving. Skipping the sample means the first time the garment exists is also the first time it is made in quantity, so every construction problem is discovered at production scale. Brands do it on simple repeats where the pattern, fabric and construction are all unchanged. On a new style it is usually the most expensive economy available.

Can you work from a pattern we already own?

Yes, and it is often the fastest route because the fit work is already done. Send the pattern with its size set and, if possible, the fit-approved garment. What we need to check is that the pattern's grade rules and the fabric we are using behave the way the original did — a pattern developed in a stable woven cloth can behave differently in a fluid one.

Do you manufacture from body measurements or from a size chart?

From a size chart with point-of-measure tolerances, which is what makes a run consistent. Body measurements are useful for fitting a sample on a real person, but they cannot be inspected against, because two garments can fit the same body differently. A published size chart with tolerances is what allows a fault to be a number.

Will you copy a garment we send you?

We will develop from a reference garment, which is a different thing from copying it. A reference gives us silhouette, proportion and construction logic to work from; the pattern, the measurements and the final construction are developed for you. We do not reproduce another brand's product as a replica, and we would rather tell you that at the start than argue about it later.

What most often goes wrong on a first order?

Fabric and decision timing, in that order. Fabric because it is the one input nobody controls, and decision timing because comments and approvals accumulate days that nobody notices until the delivery date is close. Manufacturing faults are a distant third — the line is generally the most predictable part of the process.

Should everything be made in one factory?

For one style, yes — splitting a style splits the retained standards and the two versions drift apart. Across categories, a specialist is often better: a factory strong on structured eveningwear is not automatically strong on knitwear. The general principle is to split by capability rather than by convenience.

We are new to this. Where should we start?

With the brief and the fabric, before the pattern. Most first-time delays are not caused by inexperience with manufacturing; they are caused by a design that has not yet been reduced to decisions. Choosing a manufacturer covers the supplier side of the same question.

Conclusion

Manufacturing is best understood as a sequence of decisions, each of which converts flexibility into certainty. The brief and the pattern are cheap to change. The fabric booking and the marker are not. Between those two points sits almost all of the risk in a production project.

A brand that brings a complete brief, a fabric decision and a consolidated comment round will run inside the published windows. One that does not will still get the garment — it will simply have paid for the missing decisions in calendar time, at the point where calendar time is most expensive.