Handbag Maisonby QUANZHOU JUNYUAN BAGS
Get a quote

Craftsmanship Questions to Ask a Luxury Handbag Partner: Seventeen That End in a Number

The craftsmanship questions worth asking a luxury handbag partner are the ones that end in a number, and the order matters: hide grading first, then pitch at 8-9 per inch held to one stitch of variance over 25.4 mm, then edge film at 0.10-0.15 mm dry, then hardware seated within 0.2 mm. Ask them in writing, because a verbal answer to a measurement question cannot be re-checked at goods-in and therefore cannot be enforced.

steel rule, loupe and stitch gauge laid beside a clicked leather panel and a folded pattern piece
A twenty-minute call covers far more ground when every question ends in a figure that can be re-checked at goods-in.

What Each Craftsmanship Question Is Worth Before You Ask It

A question earns its place on a call only if a wrong answer changes what you do next. Ranked by the bench minutes they protect, the hide questions come first, the seam questions second, the edge third and the hardware fourth, and that ranking is not a matter of taste: it follows the cost of putting each one right after the pieces are already cut and closed. Ask them in writing and ask for a number back, because a verbal assurance about pitch or film thickness cannot be checked at goods-in and therefore cannot be enforced.

Five questions in the order they should be asked, with the bench cost of leaving each one unanswered.
QuestionWhat it protectsCost when it goes unansweredTime to verify
How is the hide graded and nested?Grain direction, panel yield, colour match across panelsRe-cut of 8-12 per cent of the panel set6 min per hide
What pitch and thread are specified?Seam strength, slant consistency, repairabilitySeam rework on 15-20 pieces per 5004 min per seam
How many coats on the edge?Film thickness, adhesion at the corner mitreEdge re-coat at 8-12 min per piece3 min per piece
How is hardware seated and plated?Plate gap, plating thickness, salt spray resultHardware reset at 3-5 min per piece5 min per piece
How is the run held to the sample?Consistency across all 500 piecesSecond inspection and a re-work reserve20 min per five pieces

Notice what the table assumes: that every question ends in a figure. Asking whether care is taken with the edge is not a question, it is an invitation. Asking how many coats are applied, at what dry film, and with what flash-off between them, is a question, and the answer can be measured on the second piece of the run as easily as on the first.

Questions About the Hide Before the Knife Goes Down

Start here, because grade placement decides most of what a finished piece looks like and nothing downstream can repair it. Ask how the hide is graded, how many grades the panel set is drawn from, and where each panel sits in relation to the spine and the belly. Ask for the thickness band the shell is split to, typically 1.2-1.4 mm for a structured body and 0.9-1.1 mm for a soft one, and ask which hide lot the panels come from, because a run split across two lots will show two colours in the same photograph.

  • Which grades of the hide are used for which panel, and what is rejected outright?
  • Is the shell split to a stated band, and how is that band checked on a delivery?
  • What minimum finish adhesion is held, and under which test method?
  • Are panels nested for grain direction, or for yield alone?
  • If the run spans two hide lots, how is the colour match evidenced?

The last of these is the one most often skipped and the one that costs most when it fails. Two lots sharing a reference can sit a half-shade apart, and on a piece with one large uninterrupted panel that difference is visible from two metres. Insist on panels drawn from a single lot per colourway, or on a blocking plan that groups every panel by lot before the knife goes down.

Questions About Pitch, Thread and Seam Geometry

Stitch pitch, machine
8-9 per inch, 2.82-3.18 mm between holes
Pitch variance
One stitch over 25.4 mm, read at three points
Thread, machine
Bonded nylon Tex 40, 0.45 mm
Thread, hand
Waxed linen 18/3, 0.55 mm
Slant
15-20 degrees, constant along the run
Seam allowance
10 mm on body, 8 mm on facing
Seam line deviation
Under 0.5 mm over a 100 mm run
Back-tack
3 stitches, buried, not showing on the face

Pitch and thread belong together because they interact. A bonded nylon at Tex 40 does not behave like a waxed linen at 18/3, and a bench that changes thread between sampling and bulk changes the feel of the seam even when the pitch is identical. Ask which thread is on the approved sample and ask that the same reference is written on the tech pack, with substitution permitted only after a fresh sample is approved. The trade-off between the two methods is set out in machine versus hand stitching on luxury handbags, and the hand method itself in how hand stitching elevates a luxury handbag.

Then ask the geometry questions: allowance, slant, back-tack and deviation. A 10 mm allowance on a body seam is what allows a repair to be made later without shortening the panel. A three-stitch back-tack buried under the seam rather than showing on the face is what stops an end from lifting in the second month. Deviation under 0.5 mm over a 100 mm run remains the fastest single test of whether the panels were clicked and nested with care.

Questions About Edge Work and Corner Building

Edge work is where bench time shows, and it is where a weak programme is most easily disguised by a photograph. Ask for the number of coats, the dry film each produces, the flash-off between them, and the sanding grade used before a coat goes on. Three coats at 0.10-0.15 mm dry, with 10-12 minutes of flash-off at 60 degrees and a light pass at 400 grit between coats, is the specification a designer piece should carry.

  • How many coats, and what dry film does each one produce?
  • What flash-off time and temperature are held between coats?
  • Which grit is used between coats, and is the edge sealed before or after assembly?
  • How is the corner mitre closed, and what gap is accepted there?
  • Is the edge re-checked after hardware setting, which can lift a fresh film?

That final question catches a common failure. Hardware set after edge work can lift a film that has not fully cured, and the damage appears as a hairline at the plate edge two weeks later, by which point the piece is boxed and shipped. Ask for hardware to be seated before the final coat, or for a stated cure time to be observed before any metal is set against a finished edge.

Questions About Structure, Interlining and the Base

01

Interlining weight

0.8 mm on a structured body panel and 0.4 mm where the panel has to turn. Heavier is not finer; it reads as stiffness and the piece creases rather than softening with use.

02

Base board and feet

A 1.5 mm board with five feet at 5 mm, set after edge work so the piece stands level when empty. Ask where the board sits in the stack, not only whether there is one.

03

Handle drop

Measured with the bag empty under a 250 g load, with the two sides agreeing within 2 mm. This is the measurement a customer reads first when the piece is carried.

Structure questions come last in this group because they are the ones a sample conceals best. A piece can look crisp on a form and slump in the hand by the second week, and the difference is usually interlining weight and the position of the board in the stack rather than the shell itself. Ask for the stack order in writing, from the outer shell inward, and ask what happens to it when the lining is inserted.

Questions About Hardware, Plating and Closure Feel

Three things are asked of every metal item: plating thickness, salt spray result and seating tolerance. On brass, 0.15-0.3 µm is the working band, verified on a sample from each lot rather than once a season, because a plating bath drifts within a week. Salt spray is run for 24 hours on a standard specification and 48 hours on a high-wear one, and the result is recorded against the lot rather than the order.

  • What plating thickness is specified, and is it checked per lot or per season?
  • How many hours of salt spray does the finish specification carry?
  • What gap between plate and panel is accepted, 0.2 mm or more?
  • How does the closure feel at sample stage, and is that feel written down?
  • Are zips specified by size and tape reveal, or only by colour?

The fourth question is rarely asked and often decisive. Closure feel is a specification: a turn-lock should close with resistance and release without force, and that resistance comes from a setting adjustment rather than from the metal itself. If it is not written down at sampling, it will not be reproduced at 500 pieces, and the difference will be read by customers as a difference in quality rather than as a difference in setting.

brass turn-lock plate seated flush against a black leather panel, photographed at a low angle
Seating within 0.2 mm, plating at 0.15-0.3 µm, and a closure feel written down at sampling rather than discovered at goods-in.

Questions About Holding the Run to the Approved Sample

Every question above describes one piece. This one describes 500. Ask what happens between the approved sample and the last piece of the run: which measurements are re-checked, how often, by whom, and what the re-work reserve is. An inspection drawn at AQL 2.5 under ISO 2859-1 gives a defined sample size with an accept and a reject number, which is what turns a claim of careful checking into a figure that can be argued about and then enforced.

The question behind the question

Ask what the bench does when a measurement falls outside tolerance at piece 300. A programme that answers that well will answer every other craftsmanship question well too.

Also ask for the record. Readings rather than verdicts, taken on five pieces rather than one, kept against the order so that a second drop can be compared with the first. Consistency across a run is the hardest thing to buy, and a programme that records it is a programme that can repeat it on the next colourway.

Questions About Sampling Windows, Minimums and Shipping Terms

Craft questions end in commercial ones, and these belong on the same call. Sampling takes 6-10 working days on a standard build and 12-15 on a complex one; bulk occupies 35-50 days; minimums sit at 500 pcs per style and colourway; and shipments leave FOB Xiamen, with FOB, CIF and DDP terms all available. Tightening edge build-up and hardware seating from an ordinary tolerance to the figures named above adds roughly USD 4.80 per unit.

The programme runs through our validated partner factory Quanzhou Yingming Bags, an SGS-validated site with 7 production lines and about 200,000 pcs per month of capacity, holding export experience since 2014. Inspection is drawn at AQL 2.5 under ISO 2859-1 with a re-work reserve held, and material is tested alongside it against ISO 3376 for tensile strength, ISO 11644 for finish adhesion and ISO 105 for colour fastness. Hide traceability is evidenced through Leather Working Group tannery audits, and where an exotic shell is specified, CITES documentation is checked before a panel is released.

Numbers to have in writing before the call ends

  • Pitch at 8-9 per inch with one stitch of variance over 25.4 mm; seam deviation under 0.5 mm per 100 mm; edge film 0.10-0.15 mm dry after three coats; hardware gap under 0.2 mm; handle drop within 2 mm; sampling 6-10 working days and 12-15 when complex; bulk 35-50 days; 500 pcs per style and colourway; AQL 2.5 under ISO 2859-1; FOB Xiamen.

Two of those deserve a sentence of their own. The minimum is per colourway and not per order, which is the detail that most often surprises a first programme; and the sampling window runs in working days from an approved tech pack rather than from an enquiry, so a week lost in specification is a week lost twice. The bench itself is described on the craftsmanship page, and the way the plan is drawn and recorded on luxury quality control.

A Twenty-Minute Call Script, In This Order

  1. Open with the hide: grade, thickness band, single lot per colourway.
  2. Move to the seam: pitch, thread reference, allowance, back-tack.
  3. Then the edge: coats, dry film, flash-off, grit, mitre gap.
  4. Then structure: interlining weight, board position, handle drop under load.
  5. Then hardware: plating thickness, salt spray hours, seating gap, closure feel.
  6. Then the run: how the sample is held, what is recorded, what the reserve covers.
  7. Close with timing and terms, and read every figure back in writing.

Twenty minutes is enough, and the order matters more than the duration. Questions asked in this sequence build on one another, so a vague answer at step two is exposed by a precise answer at step four. Asked in reverse, the same questions produce a set of assurances and no information, which is exactly the outcome the script is designed to avoid.

Answers That Should End the Conversation

Three answers are disqualifying, and none of them is rude to hear. A reassurance offered in place of a number. A tolerance deferred to the sample. And a re-work reserve described as unnecessary. Each transfers a decision that belongs in the specification into a conversation that happens after the pieces are cut, which is the most expensive possible place to hold it.

  • A number replaced by a reassurance on any measurement question.
  • A refusal to write tolerances onto the tech pack before sampling begins.
  • A plating or salt spray result described as typical rather than specified.
  • No answer to what happens when piece 300 falls outside tolerance.
  • A sampling window quoted from enquiry rather than from an approved tech pack.

Everything else is negotiable, and a good partner negotiates it openly: which tolerances apply where, how many pieces carry the full set of checks, and what the re-work reserve covers. The point of the questions is not to find a perfect bench. It is to find one that will say, in numbers, what it intends to do and then be held to it.

Send the questions and get the numbers back

Send your drawings, material direction, quantity per colourway and delivery date. You will receive written answers to every measurement question above, naming the tolerance held on each. Request a quote.

Frequently Asked Questions

What is the first craftsmanship question to ask a luxury handbag partner?

Ask how the hide is graded and nested, because grade placement decides most of what the finished piece looks like and nothing downstream can repair it. Follow it with the thickness band the shell is split to, 1.2-1.4 mm on a structured body, and which hide lot the panels are drawn from.

How many stitches per inch should I ask for?

Eight to nine on a body seam, which is 2.82 to 3.18 mm between holes, with no more than one stitch of variance over 25.4 mm read at three points. Twelve per inch is used for fine detail but begins to perforate a 1.2-1.4 mm shell rather than sew it.

Which thread should be written onto the tech pack?

Name the reference, not the type. A bonded nylon at Tex 40 measures 0.45 mm and behaves differently from a waxed linen at 18/3 measuring 0.55 mm. Whichever is on the approved sample should be written down, with substitution allowed only after a fresh sample.

What edge specification should I ask for?

Three coats producing a dry film of 0.10-0.15 mm, with 10-12 minutes of flash-off at 60 degrees between coats and a light pass at 400 grit. Ask also whether the edge is sealed before or after assembly, and whether it is re-checked once hardware is seated.

How do I ask about hardware without sounding overly technical?

Ask three plain things: how thick is the plating, how many hours of salt spray does the finish carry, and how large a gap between plate and panel is accepted. The answers to look for are 0.15-0.3 µm, 24 hours standard and 48 hours for high wear, and 0.2 mm.

What should the interlining weigh on a structured body?

Around 0.8 mm on a body panel and 0.4 mm where the panel has to turn. Heavier is not finer: a thick interlining reads as stiffness and the piece creases rather than softening with use. Ask for the stack order in writing, from the outer shell inward.

How is handle drop measured and what tolerance should I ask for?

With the bag empty and a 250 g load applied, the two sides should agree within 2 mm. Beyond about 4 mm of difference the piece reads as hanging crooked, and it is the first thing a customer notices when the bag is carried rather than photographed.

What does AQL 2.5 mean for a run of 500 pieces?

It sets the sample size drawn from the run and the accept and reject numbers that go with it, under ISO 2859-1. In practice it converts a claim of careful checking into a defined plan with a re-work reserve, so a finding is corrected on the bench rather than at the destination.

How long should sampling take?

Six to ten working days on a standard build and twelve to fifteen on a complex one, counted from an approved tech pack rather than from an enquiry. That distinction matters: a week lost in specification is a week lost twice over.

What is the minimum order per colourway?

500 pcs per style and colourway. The figure is per colourway rather than per order, which is the detail that most often surprises a first programme, because three colourways of one style sit at 1500 pieces in total.

Which answers should end the conversation?

A reassurance offered in place of a number, a tolerance deferred to the sample, and a re-work reserve described as unnecessary. Each moves a decision out of the specification and into a conversation held after the pieces are already cut.

How do I make sure the answers are binding?

Ask for them in writing on the tech pack, with each tolerance named against a check and a frequency. A verbal answer to a measurement question cannot be re-checked at goods-in, and what cannot be re-checked cannot be enforced later.

People Also Ask

Which single question reveals the most about a handbag partner?

Ask what happens when a measurement falls outside tolerance at piece 300. A programme that answers that precisely will answer every other question precisely; one that answers it vaguely will not hold a tolerance anywhere.

Can tolerances be tightened after sampling?

They can, but at a cost in bench time and a second sampling round. Tightening edge build-up and hardware seating alone adds roughly USD 4.80 per unit, which is why tolerances belong on the tech pack before the first sample.

How many pieces should be checked from a 500-piece run?

Five taken by the AQL 2.5 plan gives a defensible read, and readings rather than verdicts should be recorded against the order so a later drop can be compared with the first.

Is a thicker interlining a sign of better craft?

No. It usually reads as stiffness and produces a piece that creases rather than softening. Structure comes from the board and its position in the stack, not from the weight of the interlining alone.

Why ask about the hide lot rather than the hide type?

Because two lots under the same reference can sit a half-shade apart, and on a large uninterrupted panel that difference is visible from two metres. One lot per colourway removes the risk entirely.

Sources and further reading

Every figure on this page is either our own programme data or attributed below. Open the originals before you write them into a specification.