How to Approve a Luxury Handbag Sample Correctly

What an approval actually commits you to

A signed sample is the reference against which bulk will be inspected. Every dimension on it becomes the target, every tolerance you did not specify becomes the tolerance on the piece in front of you, and every fault you accepted becomes acceptable in production. That is a great deal of authority to place in one object, which is why the assessment has to be systematic rather than impressionistic.

It also commits the other side. Once a sample is signed, material is called off, the line is scheduled and the bulk window begins. A change requested after that point is not a correction, it is a re-specification, and it carries the cost and the delay of one.

Before you assess: the five conditions

Assessment should not begin until five things are true. The pack is complete and carries a revision number. The correction list from the previous round is available, so each point can be confirmed closed. The signed colour reference panel is present in the room. The piece has rested for at least a day after finishing, because edge paint and adhesives continue to settle. And one person is present with authority to decide.

• The pack is complete, versioned, and matches the piece being assessed.
• The previous round's correction list is to hand, numbered and unanswered points highlighted.
• The signed colour reference panel is in the room, not in a file elsewhere.
• The piece has rested at least a day since finishing was completed.
• One person is present who can decide, rather than a group who can only comment.

The checklist, part one: external

Start outside and work in. Look at the piece under three illuminants in a light booth, because a fault that is invisible under daylight can be obvious under retail fluorescent. Then work through the surfaces systematically rather than in the order your eye happens to take.

1. View the piece under D65, TL84 and incandescent in sequence and note any metamerism.
2. Check panel alignment at every seam: flap to body, gusset to front, pocket to panel, within 0.5 mm.
3. Count stitch density on visible topstitching: 8 stitches per inch unless the pack states otherwise.
4. Check the stitch line for consistency of distance from the edge and for any skipped or doubled stitch.
5. Run a hand along every edge for build continuity, smoothness and even sheen against the reference chip.
6. Check that the bag stands squarely on a flat surface and that all feet touch together.
7. Check symmetry: handles matched, straps even, hardware centred on its stated datum.
8. Confirm surface cleanliness: no adhesive residue, no tool marks, no stray pen or gauge marks.

The checklist, part two: internal and function

The inside of a bag is where a client forms a private judgement, and it is where assessments are most often cursory. Open every pocket, run a hand along every lining seam, and check that the lining is anchored rather than floating. Then use the bag: load it with a realistic weight and carry it, because several faults only appear under load.

• Open and close every pocket and check that each is anchored, not floating inside the body.
• Run a hand along every lining seam for tension consistency and any open stitch.
• Check the lining for clean corners, no puckering at the base and no visible adhesive.
• Load the bag with a realistic weight and carry it, checking handle comfort and strap behaviour.
• Check that the bag holds its shape loaded, and that the closure still engages without strain.
• Check that the interior volume is usable and that nothing intrudes into it unexpectedly.

The checklist, part three: hardware and closure

Hardware is assessed in two ways at approval: by hand on the piece, and by record from the component testing that preceded it. The hand check catches seating and feel; the record catches whether the component itself passed its regime. Both are needed, because a closure can pass component testing and still be poorly set.

1. Cycle every closure twenty times by hand and note any change in resistance or sound.
2. Check that every plate seats flatly with no rocking, and that escutcheons centre within 0.5 mm.
3. Check plating shade against the reference clasp under D65 and TL84.
4. Confirm the component record: 0.5 microns plating minimum, 24 hours salt spray, 5,000 cycles, 25 kg pull.
5. Check that feet, rivets and anchors are set evenly and that no leather is distorted around them.
6. Check that magnetic closures separate at the specified force and do not open unintentionally.

The checklist, part four: colour and measurement

Colour is measured, not judged. Place the production panel beside the signed reference on a neutral grey card, view at about 60 centimetres under D65, and read the Delta E. Our standing limits are 1.0 or better for a body panel against the reference and 0.8 between adjacent panels on one bag. Then measure the piece against every dimension in the pack and record the deviation.

What to record at approval

An approval is a set of records, not a handshake. Capture the measurement sheet with every dimension and its deviation, photographs of the piece under each illuminant including close views of hardware and edges, the colour reading, the component test records, and the signed sample itself. The record should let someone a year later reconstruct exactly what was approved without asking anyone.

Where an approval is given with a noted exception — a detail accepted despite deviating — that exception is written down with the reason. Unrecorded exceptions are how a bulk order inherits a fault nobody agreed to, and how an argument starts at final inspection months later.

How to decline a sample

Declining is straightforward when it is evidence-based. Number the findings, classify each as critical, major or minor, and state which are structural and which are local. Give the piece back with the list rather than with an impression, and say what would make it approvable. A decline that names the correction is a request; a decline that says the piece feels wrong is an argument.

It is also worth being clear about the difference between declining and re-specifying. Declining asks for the pack to be met; re-specifying changes the pack. Both are legitimate, but they carry different costs and different calendars, and conflating them is how a programme loses weeks while believing it is in correction.

After approval: sealing the reference

Once signed, the sample is dated, photographed and stored flat, away from light, in a sealed sleeve. It travels nowhere except to the bench when a question arises. A second piece, where one exists, is kept as a working reference for inspection, and the two are compared periodically so that drift in the working reference is caught before it causes a wrong decision.

A correction made after approval costs more than one made before it: re-cutting and re-finishing one panel costs around USD 4.80 in bench time on a sample, and the same correction made after bulk has been cut costs the run. That asymmetry is the reason the checklist exists, and the reason it should be worked through slowly rather than skimmed.

Who should be in the room

Approval goes better with three roles present and only one decision-maker. Someone from the design side to judge intent, someone from the bench to explain what is physically achievable, and one person who decides. Committees do not approve well, because each member optimises for a different outcome: design for intent, production for feasibility, buying for date. Those tensions are legitimate and should be voiced, but they have to be resolved by one person at the end of the session rather than left open.

Where the decision-maker cannot attend in person, we send measured photographs under a named illuminant and a completed measurement sheet rather than asking them to judge from a video call. Colour and texture do not survive compression, and an approval given over a call is an approval of an image.

Working through the checklist in one session

The full twenty-four points take about ninety minutes on a structured bag, and that is time well spent. Work in the order given: external, internal, hardware, colour, measurement, then records. Do not skip to the parts you care about, because the parts nobody checks are the ones that become the standard. Where a point cannot be assessed — no light booth available, no colour instrument to hand — say so and defer rather than guessing.

Checklist points

24 across external, internal, hardware, colour and measurement

Assessment time

About 90 minutes on a structured bag

Illuminants

D65, TL84 and incandescent, viewed in sequence

Body panel shade

Delta E 1.0 or better against the signed panel under D65

Adjacent panels

Delta E 0.8 or better on the same piece

Panel alignment

Within 0.5 mm at every seam

Stitch density

8 stitches per inch on visible calfskin topstitching

Hardware position

±0.5 mm from the stated datum

Closure cycling

20 cycles by hand at approval; 5,000 on a rigged frame per style

Component record

0.5 microns plating, 24 hours salt spray, 25 kg pull

Correction cost after approval

Around USD 4.80 per re-cut and re-finished panel

Minimum order quantity

500 pieces per style per colourway

Sampling lead time

6 to 10 working days, 12 to 15 for complex constructions

Bulk lead time

35 to 50 days from approved sample

Shipment terms

FOB Xiamen

Frequently Asked Questions

People Also Ask