The Wholesale Handbag Quality Grading System Buyers Should Know

Inspection reference points

Question 1: Is There a Formal Grading System?

Not in the sense buyers usually mean. There is no universal stamp that declares a handbag grade A, and any seller using the term without a written definition is using it loosely. What does exist is a rigorous, internationally recognised system for deciding how many pieces to examine and how many defects are tolerable: ISO 2859-1, which defines sampling procedures and accept-reject numbers by lot size and inspection level.

That distinction matters commercially. A buyer who asks for grade A goods is asking for an opinion; a buyer who asks for inspection at AQL 2.5 under ISO 2859-1 with critical defects at zero is asking for a measurement with a defined outcome. The second can be written into a contract and a payment condition. The first cannot. Everything in this guide follows from that substitution, and it is the single most useful change a buyer can make to how quality is discussed.

The second thing worth knowing is that sampling is statistical, not exhaustive. An 80-piece sample from a 500-piece lot is a statement about confidence, not a statement about the 420 pieces that were never opened. That is not a weakness of the system; it is the design. The buyer's protection comes from what happens when the sample fails, which is question five. The wholesale handbag programme guide sets out where inspection sits in the order cycle.

Question 2: How Many Pieces Are Examined?

The number comes from two inputs: the lot size and the inspection level. At general inspection level II, which is the default for finished goods, a lot of 501-1,200 pieces draws code letter J and a sample of 80, with accept 5 and reject 6 at AQL 2.5. A lot of 1,201-3,200 draws code letter K and 125 pieces, accept 7 and reject 8. A lot of 3,201-10,000 draws code letter L and 200 pieces, accept 10 and reject 11. Those three bands cover almost every handbag order placed at this level.

Two further choices sit behind those numbers and both should be stated in writing. The first is the inspection level: level II is the default, level I draws a smaller sample and level III a larger one, and a buyer who wants more assurance asks for level III. The second is the AQL itself: 2.5 is the standard for major defects on finished bags in this trade, with 1.0 used where the buyer wants tighter control and 4.0 commonly applied to minor defects. Critical defects are separate from all of these and are held at zero.

Question 3: How Are Defects Classified?

Three classes, and the classification does more work than the count. A critical defect is one that makes the product unsafe or non-compliant — a broken strap root that could drop the bag, exposed sharp hardware, a coating that fails a restricted-substance limit. A major defect is one a customer would notice and reject: visible seam grin, colour off the approved swatch beyond tolerance, a zip that will not run, a logo in the wrong place. A minor defect is one most customers would not notice and would not return over: a loose thread under 5 mm, light creasing from packing, a small mark on the lining.

Critical defects are held at zero on any credible programme, which means a single one fails the lot regardless of what the AQL allows elsewhere. Major defects are typically inspected at 2.5 and minor at 4.0. That split is what makes the system workable: a lot with four minor defects and no majors is a good lot, and a lot with two majors is not, even though the raw counts are similar. Without the classification, the counts are meaningless.

Question 4: Major or Minor?

How to judge major from minor

The boundary is the customer's reaction, not the inspector's opinion, and that is why it belongs on the spec sheet. The working test is whether a reasonable customer would return the bag on seeing it. A 3 mm shade difference between the body and the strap is a minor if it is only visible under studio lighting and a major if it is visible in a shop. A stitch skip is a minor on a lining seam and a major on a load-bearing handle join. The same defect changes class with position.

• Critical: broken strap root, exposed sharp edge, non-compliant coating, any safety or compliance failure.
• Major: seam grin, colour beyond the agreed tolerance, zip failure, logo misplacement, lining tear, hardware tarnish.
• Minor: loose thread under 5 mm, light creasing from packing, small ink mark on the lining, minor asymmetry under 3 mm.
• Position changes class: a stitch skip is major on a handle join and minor on a lining seam.
• Write the boundary down: a defect list agreed before production removes the argument afterwards.

The measurement thresholds are what convert opinion into a check. A colour tolerance of roughly ΔE 1.5 against the approved swatch, a loose thread limit of 5 mm, an asymmetry limit of 3 mm, a stitch density of 8 per inch on a #18 needle, and a bar-tack of roughly 6,000 stitches at each handle root. Each of those is a number an inspector can measure, and each one removes a class of argument. ASTM test methods exist for many of these properties where a buyer wants a formal test rather than a visual judgement.

Question 5: What Happens When a Lot Fails?

What failure actually means

Failure has a defined meaning: the number of defects in the sample meets or exceeds the reject number for its class. What follows should be agreed before production, because it is the part of the system that actually protects the buyer. The normal ladder is: a 100% sort of the lot at the programme's cost with re-inspection; rework of the specific defect if it is reworkable; or rejection with replacement or credit. Which applies depends on the defect, and a specification that does not say so leaves the decision to whoever is holding the goods.

Timing matters as much as remedy. A lot inspected at 80% packed and found failing can be sorted before the container is sealed; the same lot inspected after sealing has become a claim at destination, with freight, storage and a season at stake. That is the practical argument for inspecting during production as well as at the end: a during-production check finds a systematic defect while there is still time to correct the remaining pieces rather than after all of them are made.

There is also a commercial decision the buyer makes at this point, and it is often the right one: accept the lot with an agreed allowance. Where the defect is minor, where the season is close, and where the count is only marginally over, a credit against the next order is frequently better business than a replacement that arrives after the window. The returns and defect claim procedure sets out how that is documented.

Question 6: Where Do A-Grade and B-Grade Fit?

Informal grading terms are useful shorthand and dangerous as specification. A-grade usually means goods that pass the agreed inspection; B-grade usually means goods with visible but acceptable faults, sold at a discount; and C-grade usually means goods that failed. None of those definitions are standard, and a buyer who accepts a grade without a defect list behind it has accepted nothing measurable. Ask what the grade means in defect counts, or replace it with an AQL.

Where grading is genuinely useful is in the handling of the tail. A run that produces 500 good pieces and 20 with a minor hardware blemish can be sold as two tiers: 480 at full price and 20 at a discount, rather than as a failed lot. That requires the blemish to be defined and counted, not merely noticed. Agreeing a minor-defect allowance at the outset — often 0.5-1% of shipped quantity — makes that split possible without a dispute and gives both sides a number to work to.

How Grading Affects Where You Can Sell

The inspection level a buyer chooses determines which channels are open to them. A marketplace with a generous returns policy effectively imposes its own quality standard through the return rate: goods that pass AQL 2.5 at the factory can still generate an 8-12% return rate driven by fit and expectation, but goods with unclassified defects generate one driven by disappointment, which is higher and far more damaging to a listing. Department and boutique wholesale, by contrast, usually imposes a documented standard of its own and will ask for the inspection report before listing.

Compliance is the harder gate and it does not grade — it passes or fails. A REACH declaration against Annex XVII for the EU, a Prop 65 assessment for California and CPSIA documentation where an item could be read as a children's product are binary, and a shipment without them can be held regardless of how good the goods are. That is the practical reason to treat compliance as a critical defect class with zero acceptance rather than as paperwork attached at the end.

The Defect List Worth Writing Down

That table is not exhaustive, but it covers the defects that account for most inspection outcomes on a handbag run, and it shows the pattern: every line has a measurement and a remedy. A defect list without those two columns is a description rather than a specification. Agreeing it before production is the cheapest quality investment available, because it converts every later conversation from an argument into a count. The operational sequence around it is set out in the first-time buyer checklist, and the recovery route once something has shipped is in returns and defect claims.

Building Quality Into the Specification

Inspection finds defects; specification prevents them, and the two are not substitutes. A handle root bar-tacked at roughly 6,000 stitches does not fail; a colour tolerance written at about ΔE 1.5 does not become an argument; a hardware finish specified with a 24-hour salt spray result does not tarnish in a customer's wardrobe in month two. Every one of those is a decision made before production rather than a check made after it, and together they remove most of the defects an inspection would otherwise find. The place those decisions get tested is the sampling sequence before bulk, where a written threshold becomes a physical object a buyer can hold and measure.

The reverse is also true and worth stating plainly. A buyer who specifies loosely and inspects tightly creates a system that fails good lots and passes bad ones, because the inspector is measuring against a spec that does not say what matters. The efficient combination is a tight specification and a standard inspection: AQL 2.5 under ISO 2859-1 with critical defects at zero, applied to a defect list that was written before the run began.

What an Inspection Report Should Contain

The six lines every report should carry

Before the balance is released, the report should show six things: the lot size and the sample size with its code letter, the AQL applied to each defect class, the count found in each class, the accept and reject numbers used, photographs of every defect recorded, and a clear pass or fail with the inspector's name and date. A report missing any of those is not evidence, and a balance paid against a report that does not state its own sample size is a balance paid on trust.

Two further items turn the report into protection. The first is the inspection timing — whether it was performed at 80% packed, at 100% packed, or during production — because that determines what could still be corrected. The second is who performed it: a third-party inspection can be arranged if the buyer prefers a report not issued by the line that made the goods. Neither costs much, and both change what the report is worth when it is needed.

Send the silhouette and the quantity, and a private label coordinator will confirm the inspection level, the sample size that applies to your lot, and the defect list the run will be measured against. Get a quote.

Calibrating the Inspector, Not Just the Gauge

Two inspectors handed the same specification and the same lot will not return the same defect count, and the gap is wider than most buyers expect. Calibration narrows it: a retained reference board of accepted and rejected examples, a fixed light source, and a round-robin count on the same twenty units before each inspection. Where two inspectors agree on fewer than about 85 per cent of units, the number in the report is not yet trustworthy.

• Reference board: about 12-20 retained units, each labelled with its class and the measurement that decided it.
• Light source: D65 at a stated distance, plus a second check under the destination's store lighting.
• Round-robin: two inspectors count the same 20 units; agreement below about 85 per cent means recalibrate first.
• Drift: re-run the round-robin every three lots or every quarter, whichever comes first.

Switching Rules: What Happens After the First Few Lots

Inspection is not a fixed gate; it moves. The switching rules in ISO 2859-1 move a programme between normal, tightened and reduced inspection on the recent history of lot results, and the consequence is commercial rather than statistical. Tightened inspection draws a larger sample against a lower accept number, so more lots fail at exactly the moment a calendar is least able to absorb it.

Frequently Asked Questions

People Also Ask