On this page
- The approved sample controls only what was actually approved
- Five samples, five different jobs
- 1. Sample fabric and bulk fabric are not always the same production lot
- 2. Patterns can change after the sample is approved
- 3. One sample machinist is not the same as a production line
- 4. Tolerances may be too loose for the silhouette
- 5. Trims that look similar can perform differently
- 6. Pressing and finishing change the final garment
- 7. Packing can undo production quality
- The approval sequence that protects bulk production
- When bulk comes back off-standard
- Keep a sealed-sample library
- A practical brand checklist before cutting
- Frequently asked questions
- Working With Luxudress
- Related pages

An approved sample feels like a finished decision. The silhouette works, the fabric falls correctly, the zipper closes cleanly and the measurements are within tolerance. A brand approves it, books production and expects every bulk garment to reproduce the same result.
That expectation is reasonable, but it is not automatic. A sample is one garment made by the most experienced machinist in the building. Bulk production is a system: fabric across several lots, graded pieces cut in stacks, operators repeating one operation each, then pressing, inspection and packing. Unless the approved standards cross into that system, the differences accumulate.
The practical question is not whether variation exists, because some always does. It is which variables must be locked before cutting, how tightly each must be held, and how the factory and the brand will measure the result.
The approved sample controls only what was actually approved
A sample cannot control an unstated decision. If the brand approves the appearance but never records the fabric code, colour standard, measurements, seam construction, trim reference and finishing method, everyone downstream reads the approval differently — reasonably, because the document does not settle the question.
The most useful approval package contains four connected standards:
Scroll to see all columns
| Approved standard | What it controls | What happens if it is missing |
|---|---|---|
| Sealed sample | Overall appearance and workmanship | Bulk teams rely on photographs or memory |
| Final specification sheet | Measurements and tolerances | Fit drifts during grading and sewing |
| Approved material and trim references | Hand feel, colour, stretch and component quality | Substitutions appear similar but behave differently |
| Written construction notes | Internal method and finishing | Operators use the easiest familiar method |
The physical garment matters because photographs cannot carry hand feel, pressure, stretch recovery or internal support. The written specification matters because a garment alone does not state tolerance, stitch density or grading rule. Neither substitutes for the other.
The failure this section exists to describe is approval by conversation. A fit meeting is held, everyone agrees the sample looks better, somebody says go ahead with that, and nothing is written down. Three weeks later the line has a garment and no document says which of the two was approved.
Five samples, five different jobs
Brands say the sample as if there were one. There are usually several, and each answers a different question. Confusing them is the most common reason a brand approves one thing and receives another.
Scroll to see all columns
| Sample | The question it answers | What it cannot tell you |
|---|---|---|
| Development sample | Can the design be built at all, and does the proportion read as intended | Nothing reliable about fit in a graded range |
| Fit sample | Does this size fit the intended body, and does the construction sit correctly | Nothing about the sizes not cut |
| Size set | Do the grade rules hold across the range, and do the extremes behave | Nothing about bulk material if unapproved |
| Pre-production (PP) sample | Does the garment match approval in actual bulk materials and pattern | Nothing about the first pieces off the line |
| Top-of-production sample | Do the first pieces leaving the line match the sealed standard | Nothing about later pieces without checkpoints |
Not every style needs all five. A simple elastic-waist dress can move from development to PP with a measurement check between. A boned corset dress, a tailored blazer or a moulded cup usually needs the fit sample and the size set, because those catch failures that would otherwise surface in bulk.
The trap is treating the development sample as the fit sample. A development sample tests whether the design can exist; it is often cut from whatever is on the sample-room shelf, at a different grain, without graded pieces. Approving it as though it settled fit is how a brand ends up arguing about a waistline that was never agreed.
1. Sample fabric and bulk fabric are not always the same production lot

Many first samples are made from available development fabric. It may match the requested composition and general weight but come from a different dye lot, finish or mill run. Bulk fabric can therefore look or behave differently even when the supplier uses the same commercial name.
For satin, a small change in weight or surface finish changes shine, drape and seam behaviour. For stretch woven fabric, a change in elastane percentage or recovery can change fit. For mesh, tulle and organza, stiffness affects volume and support.
It is worth being specific, because check the fabric is not an instruction. On satin the risky variable is finish and the direction the shine reads: a lot half a shade duller shows across a whole bodice even though it passed on a swatch. On stretch woven, recovery matters more than stretch. On tulle, mesh and organza it is stiffness, because a softer lot will not hold the shape the sample held.
Before bulk cutting, confirm:
- supplier and article number;
- fibre composition;
- usable width and weight;
- stretch and recovery where relevant;
- colour standard and approved bulk lot;
- shrinkage or relaxation result;
- surface finish and hand feel.
An approved swatch is helpful, but a bulk lot approval is stronger because it evaluates the material that will actually be cut. A swatch approves colour against a standard; it does not test how the fabric cuts, seams, presses and hangs. A lot approval is a signed panel of the actual lot, large enough to drape, seam and press, compared against the approved sample.
2. Patterns can change after the sample is approved
The first sample may be made in one base size. Bulk production needs a complete size range. Grading introduces new decisions about bust, waist, hip, length, armhole, neckline and internal support, and a rule that works for a simple jersey top may not work for a corseted dress or fitted blazer.
Grading is where fit quietly leaves the building. A grade rule is a set of increments — how much the bust grows per size, the waist, the length — and those increments are a design decision, not arithmetic. A corseted dress is the hardest case, because bust, underbust and waist grow at different rates and the boning must travel with them. Grade it on a jersey bodice rule and the middle sizes fit while the ends do not.
A size set is the answer, and it is the stage brands most often try to skip. It is one garment cut in selected sizes across the range, normally the smallest, a middle and the largest, so the grade rule can be seen on a body rather than trusted on a chart. For tailoring and corsetry it is not optional.
Pattern control requires:
- a dated fit-approved base pattern;
- a written list of sample corrections;
- an approved graded measurement chart;
- a size set where the style has complex fit;
- version control so the cutting room uses the final pattern.
If a brand changes the sample after fit approval, the revised pattern and specification must carry the same revision date. Otherwise, the factory can have a correct garment and an outdated document at the same time.
Version control in the cutting room is the plain half of this and the half that causes the most damage. A pattern room can hold three revisions of one style, and the one on the cutting table is not always the one on the specification. The defence is dull: a dated pattern marked with the permitted revision, and a rule that obsolete copies are removed.
3. One sample machinist is not the same as a production line
Sample machinists are experienced at building complete garments and solving unfamiliar construction. Bulk operators repeat assigned operations. The production line needs an operation breakdown that translates the sample into repeatable steps.
Problems often appear when the sample contains techniques that are difficult to repeat at speed, such as:
- narrow satin seams that require controlled feeding;
- boning channels with inconsistent allowance;
- bias-cut panels that stretch during handling;
- hand-positioned lace or appliqué;
- invisible zippers inserted without stabilisation;
- cups, pads or internal corsetry that shift during assembly.
A pre-production meeting should identify the high-risk operations, machine settings, guides, templates and inspection points before the first production bundle reaches the line.
The operation breakdown is the bridge
The gap between sample room and line is not really skill; it is scope. A sample machinist builds the whole garment and solves each step as it arrives. An operator builds one seam several hundred times and is measured on speed and consistency. The operation breakdown converts one into the other: the sample taken apart into numbered operations, each naming its machine, stitch, seam allowance, control and check point.
4. Tolerances may be too loose for the silhouette

Two garments can both fall within a broad measurement tolerance and still look different. A fitted evening dress reacts more strongly to small waist, bust and neckline differences than an oversized shirt.
Tolerance should reflect the construction and commercial consequence. Critical measurements may require tighter control, while low-risk measurements can accept normal production variation.
The brand should identify critical points rather than assigning one tolerance to the entire garment. Examples include bust circumference, waist, neckline depth, strap length, zipper opening and finished length.
Name the points of measure rather than describe them, because waist on one garment is not waist on another. A fitted dress is measured at the narrowest point of the bodice; a drop-waist style somewhere else entirely. Naming each POM, defining how it is measured and giving it its own tolerance is the only way a bulk measurement report means anything.
Scroll to see all columns
| Point of measure | What it controls | Why it is critical |
|---|---|---|
| POM 1 bust circumference | Bodice room and cup placement | Too tight distorts the neckline; too loose gaps at the armhole |
| POM 8 underbust | Support and boning position | Sets where the corsetry sits on the body |
| POM 12 finished waist | Silhouette and closure alignment | Small differences show against the centre-back zip |
| POM 14 hip | Skirt fall and pocket opening | Decides whether the skirt skims or pulls |
| POM 20 front neckline depth | Coverage and proportion | Millimetres decide flattering versus revealing |
| POM 31 strap length | Shoulder position and support | A long strap still has to hold the bodice |
Assigning one tolerance band to the whole garment is how fit drifts, and it is also the easier document to write, which is why it happens. The cost is that the two or three measurements that decide whether the dress sells get the same slack as the ones that decide nothing.
5. Trims that look similar can perform differently
A substitute zipper, cup, hook, button, elastic or lining may appear acceptable on a table but behave differently in wear. Zipper tape stiffness affects the centre-back seam. Cup shape affects neckline fit. Lining stretch affects movement. Boning type changes support and recovery.
No trim substitution should enter bulk without a documented approval when it can affect appearance, fit, durability or care.
Trim substitution is the same problem in miniature, because a substitute is chosen to look like the original and judged on its appearance. The properties that decide the garment are almost never appearance. A zipper with the same tape colour but a stiffer tape holds the centre-back seam straighter than the approved sample did, which changes the drape the brand approved. A flexible boning swapped for rigid boning of the same width supports more and presses less comfortably.
Scroll to see all columns
| Trim | Same at approval | Behaves differently because | Where it shows |
|---|---|---|---|
| Invisible zip | Tape colour and pull | Tape stiffness and tooth height differ | Centre-back seam ride and wave |
| Moulded cup | Size marking | Dome depth and edge softness differ | Neckline gape and underbust line |
| Boning | Width and length | Rigidity and recovery differ | Waist definition and wear comfort |
| Lining | Shade on the table | Stretch and slip differ | Skirt movement and seat drape |
| Elastic | Width | Recovery and stretch differ | Fit over a full day of wear |
| Hook and eye | Finish and size | Wire gauge and tension differ | Whether the closure stays shut |
The rule is simple to state and easy to ignore: a substitution is approved against the properties the garment depends on, not against the trim’s appearance. The approval has to name those properties — tape stiffness, cup depth, boning rigidity, elastic recovery — before any substitute can be judged.
6. Pressing and finishing change the final garment

Pressing is part of garment construction. Too much heat can glaze satin, flatten texture, shrink interfacing or distort seam allowance. Insufficient pressing leaves puckering and uneven edges. Hanging time can also alter bias-cut garments before hemming.
The finishing standard should specify:
- pressing temperature appropriate to the fabric;
- use of pressing cloths or shaped equipment;
- hanging or relaxation time;
- final thread trimming and cleaning;
- garment presentation before packing.
The presser is an operator, not a housekeeper. A seam sewn correctly can be pressed into a distorted edge. On a structured dress the pressing station decides how the bodice front reads and how the hem falls.
The variables are specific: temperature against the fibre, steam against the surface finish, pressure against the seam allowance, and time — both ironing and the hanging time before the next operation. A bias-cut panel hemmed before it relaxes will be the wrong length.
7. Packing can undo production quality
A garment can pass inspection and still arrive with crushed cups, permanent folds, bent boning or damaged embellishment. The packing method must match the structure of the product and the distribution channel.
Structured dresses may require tissue support, shaped folding, protective sleeves or hanging. Beaded styles need surface protection. Long garments need a fold plan that avoids visible creases across the bodice.
Packing is the last point at which quality can be lost and the first at which the brand sees the result. The customer never sees the seam that was sewn correctly; they see the crease that crossed the bodice. A fold plan is therefore a specification: where the garment folds, which panels are cushioned, and whether it travels flat, folded or hung.
Two points are easy to underrate: the packing method should be tested on a finished garment at finished weight, and compression — not folds alone — is what deforms a cup, a bone or a beaded panel.
The approval sequence that protects bulk production
Scroll to see all columns
| Stage | Main decision | Approval evidence |
|---|---|---|
| Development sample | Construction feasibility and initial appearance | Comment sheet |
| Fit sample | Fit and proportion | Measurement report and fit comments |
| Size set | Grading across the range | Approved graded measurements |
| Lab dip / strike-off | Colour and print | Signed colour reference |
| Bulk fabric | Actual production material | Lot approval and test data |
| Pre-production sample | Final materials, pattern and construction | Sealed PP sample |
| Top-of-production sample | First pieces from the line | Production approval |
| Final inspection | Shipment conformity | Inspection report |
When bulk comes back off-standard
Sooner or later a bulk lot is found off-standard — a shade that drifted, a measurement out of tolerance, a seam a checkpoint caught late. What separates an expensive problem from a manageable one is rarely the defect itself. It is how quickly the lot is contained, how honestly the size of the problem is measured, and whether brand and factory argue about facts or preferences.
Contain first, then decide
The instinct to keep shipping while the argument continues is what turns a contained problem into a market-wide one. Hold the affected goods first and establish how many pieces are actually involved, because a defect on the inspection table and a defect across a full lot are different problems.
- Hold the affected goods and stop further shipment from the same run.
- Identify the affected units by lot, size, colour and production date, not by impression.
- Quantify how far the deviation runs — one operation, one lot, or the whole order.
- Find the cause well enough to know whether continuing production will repeat it.
- Agree the disposition with the brand before any of the goods move.
Step four is the one skipped under pressure, and it decides whether the problem is fixed or merely paused. A defect that is understood can be corrected at the operation that causes it; a defect that is only described cannot.
Sort, repair or reject
Once the size is known, each piece falls into a small number of dispositions, and the useful discipline is to decide which applies before deciding who pays. Sorting means inspecting the whole lot, not a sample. Repair brings a piece back to standard without changing what was approved. Reject means it cannot, and it does not ship.
Scroll to see all columns
| Disposition | When it applies | What it requires |
|---|---|---|
| Sort | The defect is in some pieces but not all | A full inspection, not a sample, and a clear pass/fail standard |
| Repair | The piece can be returned to the approved standard | A defined method and re-inspection |
| Reject | It cannot be returned to standard without changing approval | Removal from the shipment and a recorded reason |
| Ship with concession | Minor, understood, and accepted in writing | A written concession, not a verbal one |
The one rule worth stating plainly is that repair is re-inspected. A repaired garment that re-enters the carton unchecked is not fixed; it is a defect moved somewhere harder to find. Sorting without a written pass-and-fail standard fails the same way, because two inspectors draw the line in two places.
The commercial conversation
This is where a quality problem turns into a commercial one. Take the facts first: how many pieces, what the deviation is, and what the factory proposes. Only then does cost arise, and cost allocation depends on the cause and on what was agreed — it is not a figure the brand should invent on the call.
The conversation should produce two decisions: one about the goods and one about prevention. The goods decision is one of the dispositions above; the prevention decision is what changes at the operation so the next lot does not repeat the fault.
Keep a sealed-sample library
The standards in this article only work if the approved garment still exists, in a form nobody can misread, when the question is asked. That is the purpose of a retained, sealed sample: a physical record of what was approved, held against the style and available for comparison for as long as the style is in range.
- Seal the sample by the stage it represents, so fit, PP and top-of-production are never confused.
- Label each sealed sample with the style, the approval date and the pattern revision it was made from.
- Hold it where the factory can reach it, not only where the brand can, because the line is who needs it.
- Keep it for as long as the style is in range, including re-orders and recolours.
- Treat a lost sealed sample as a lost standard, because that is exactly what it is.
A practical brand checklist before cutting

- Is the final pattern revision dated and released?
- Does the measurement chart match that pattern revision?
- Is the bulk fabric lot approved?
- Are all visible and functional trims approved?
- Are critical measurements and tolerances identified by point of measure?
- Is the pre-production sample made in actual materials?
- Are high-risk operations documented in an operation breakdown?
- Is the packing method tested on a finished garment at finished weight?
- Is the sealed sample held by both the brand and the factory?
- Is the correction or concession method agreed in writing?
Frequently asked questions
Is an approved sample enough to start bulk production?
It can be enough only when the sample is supported by final specifications, approved materials, confirmed trims and a released pattern. The sample should be retained as the workmanship standard, and the specification should say which measurements it controls and how tightly.
Why does bulk fabric look different from the sample?
The sample may use development yardage or a different dye lot. Weight, finish, colour and stretch should be approved against the actual bulk lot rather than the swatch, on a panel of that lot under the lighting the garment will be seen in.
Should every size be sampled?
Not every style requires every size, but complex fitted garments usually benefit from a size set covering selected sizes across the range. For corsetry and tailoring, the size set is the only stage that tests the ends of the range before bulk fabric is committed.
Who approves substitutions?
The brand or authorised buyer should approve substitutions that affect appearance, fit, performance, certification or care, and the approval should name the property the substitute must match, not just the trim it replaces.
Can final inspection correct sample-to-bulk differences?
Final inspection can identify differences, but it cannot economically correct fabric, pattern or construction problems once the order is complete. Control must begin before cutting, and the inspection standard should be defined before it starts.
What is a size set, and do I need one?
A size set is one garment cut in selected sizes across the range so the grade rule can be seen on a body. If the style has a fitted bodice, a corset, a tailored shoulder or a moulded cup, a size set is the cheapest way to find a grading problem before it exists in every size.
What should be kept after approval?
A sealed sample for each approved stage, the released pattern revision, the graded measurement chart, the approved material and trim references, and the written construction and finishing notes — held against the style, so any later disagreement is settled without rebuilding the garment’s history from memory.
Working With Luxudress
Approved samples change in bulk when the approval stays a single garment instead of becoming a controlled production standard. The garment proves what is possible; the specification, released pattern, named tolerances, approved materials and operation breakdown make it repeatable.
Luxudress is the factory-direct front end for womenswear production across production facilities in Guangzhou and Dongguan, covering development, sampling, material sourcing, manufacturing, inspection and export packing.
We develop and produce structured womenswear, dresses and occasionwear through sampling, material approval, pre-production confirmation and inspection, and we hold the sealed samples and released patterns that keep a style consistent across re-orders. Where a construction detail is hard to repeat at speed, we write the operation so the line can hold it.
If you are moving a style into bulk, send the brief with the size range and the stage you are at, and we will confirm which samples and approvals it needs. You can also see how we run sampling, read how we control bulk quality, start from our tech pack guide for dresses, or follow the production process end to end.
Request a quote with the style brief, target market, size range, quantity and launch date, and we will plan the correct approval route before quotation.