AQL 2.5 in Premium Handbag Production Explained: Sample Sizes, Defect Classes and the Limits of the Standard

What AQL 2.5 Actually Means in Plain Terms

AQL 2.5 is a defect threshold agreed before production, not a defect rate measured after it: it sets how many faulty pieces in a random sample force a lot to be rejected, and nothing more than that.

The most common misreading is that AQL 2.5 means 2.5 percent of goods may be faulty. It does not. It means a sampling plan has been chosen whose operating characteristic accepts most lots at or better than that quality level, and the actual proportion of faulty pieces in an accepted lot can be higher or lower depending on where in the curve the lot sits.

Two consequences follow that matter commercially. First, a lot can pass and still contain faulty pieces, because only a sample was examined. Second, a good lot can occasionally be rejected, because sampling carries risk for both sides and the standard distributes it deliberately rather than eliminating it.

Reading the Sample Size Table

Three things fix the plan: the lot size, the inspection level and the AQL. Lot size gives a code letter, the code letter gives a sample size, and the sample size with the AQL gives the accept and reject numbers.

General Inspection Level II is the default in ISO 2859-1 and the one used on most programmes. Level I draws a smaller sample and is used where inspection is expensive or quality history is strong; Level III draws a larger sample and is used where the consequences of a miss are severe. Level II is the right default for a first season.

Read the last row of that table before agreeing anything. On a 500-piece run, four defects in fifty pieces is the line, which is tighter than most founders expect and looser than most assume once they have thought about it.

AQL 2.5 Against AQL 1.0 and AQL 4.0

Choosing an AQL is choosing how much risk the brand carries against how much the production side carries, and the three levels in common use sit at roughly 1.0, 2.5 and 4.0.

AQL 1.0 is stricter and draws the same sample but tolerates fewer defects. It is appropriate where the cost of a faulty piece reaching a customer is high relative to the cost of rework. AQL 4.0 is looser and suits goods where a minor fault is genuinely invisible in use.

2.5 is the standard setting for premium accessories because it sits where the two risks balance: tight enough that a systematic fault is caught, loose enough that a single cosmetic imperfection does not hold a season hostage.

Critical, Major and Minor: The Classification That Decides Everything

Defect classification decides more than the AQL number does, because a plan is only as good as the agreed definition of what counts: critical, major or minor, written down before bulk starts.

A critical defect is one that makes the piece unsafe or unusable — a sharp edge on hardware, a closure that will not fasten, a needle broken off inside the construction. Critical defects are held at an acceptance number of zero: one found in the sample rejects the lot regardless of any other AQL agreed.

• Critical: unsafe or unusable, held at zero acceptance, one occurrence rejects the lot.
• Major: visible or functional, counted against the agreed AQL, usually 2.5.
• Minor: a departure from specification unlikely to be noticed in use, often held at a looser AQL.
• Undocumented: the most dangerous class, because it is discovered during an argument rather than before one.

Write the classification into the plan as a defect library, not as a description. Fifteen named faults with an assigned class remove nearly all discretion from the inspection, and discretion is where most disagreements begin.

Why Handbags Are Harder to Inspect Than Garments

A handbag is a three-dimensional object with moving parts, so inspecting one takes several times longer than inspecting a flat garment and depends far more on the inspector's judgement.

Four things complicate it. Symmetry has to be judged on a turned object rather than a laid-flat one; hardware has to be operated rather than looked at; interiors have to be examined in poor light; and leather is a natural material whose acceptable variation has to be defined against a sealed reference sample rather than against a description.

Inspection time per structured bag

Commonly 3-6 minutes against well under one minute for a flat garment

Visual assessment distance

Around 30 cm for surface faults, roughly 1 m for silhouette and symmetry

Hardware operations per piece

Every closure, zip and fastening cycled, not merely examined

Reference used for shade and grain

The sealed counter sample retained at approval, opened only disputes

Critical defects tolerated in a sample

Zero, regardless of the AQL agreed for major defects

Allow for the inspection time when you plan the calendar. A 500-piece lot needs fifty pieces examined thoroughly rather than quickly, and an inspection compressed into an afternoon is an inspection that will not find what it was drawn to find.

A Defect Library: Fifteen Faults and Their Class

The library below is the practical heart of an inspection plan. Agreeing these fifteen lines before bulk starts converts an argument about quality into a count that either side can verify.

1. Closure will not fasten, or fastens insecurely — critical.
2. Sharp edge or burr on hardware that could cut — critical.
3. Broken needle or metal fragment inside the construction — critical.
4. Stitch line broken or skipped on a load-bearing seam — major.
5. Handle attachment asymmetric by more than 2 mm — major.
6. Plating scratched or blistered, visible at 30 cm — major.
7. Plating shade differs between two fittings on one piece — major.
8. Edge paint uneven, chipped, or lifted at a fold — major.
9. Adhesive residue visible on an exterior surface — major.
10. Shade or grain differs from the sealed approval sample — major.
11. Persistent odour detectable at arm's length — major.
12. Branding stamp misplaced, illegible or absent — major.
13. Panel misalignment beyond roughly 1 mm at a visible join — minor.
14. Loose thread longer than about 5 mm — minor.
15. Carton marking incorrect or carton damaged in transit — minor.

Two of these cause the most disputes: the shade comparison and the odour test, because both are sensory. The answer to both is the same — a sealed reference sample and a written protocol for how the comparison is made.

Switching Rules: Normal, Tightened and Reduced

ISO 2859-1 is not a single plan but a set of three that move: normal, tightened and reduced, with rules governing when inspection switches between them based on recent history.

Inspection starts at normal. It moves to tightened when two out of five consecutive lots are rejected, which draws a larger sample at the same AQL. It returns to normal after five consecutive lots are accepted under tightened inspection. It may move to reduced after ten consecutive accepted lots with steady production, which draws a smaller sample and accepts more risk.

Who Counts, and When

Three arrangements exist and they answer different questions: in-house inspection by the production team, a third-party inspector appointed by the brand, and a joint count at the point of packing.

Whichever arrangement is chosen, the sample must be drawn at random from finished, packed goods rather than presented. A sample that has been selected by anyone with an interest in the outcome carries no statistical meaning at all.

What AQL 2.5 Does Not Cover

AQL 2.5 covers the finished lot and says nothing about four things brands assume it covers: material compliance, durability over time, packing, and consistency between lots.

• Material compliance: chemical restrictions are answered by declarations from the tannery, not by counting faults.
• Durability: a piece can pass inspection and still fail in month three, which is why load and cycle testing matters.
• Packing: carton specification and presentation standards sit outside the sampling plan entirely.
• Lot-to-lot consistency: each lot is inspected separately, so a second delivery needs its own plan and its own reference.

This is why an inspection clause sits alongside a material clause rather than instead of one. On our programme, export experience runs since 2014 across 7 production lines with about 200,000 pcs per month of capacity, delivered through our validated partner factory Quanzhou Yingming Bags, an SGS-validated site, and inspection at AQL 2.5 is agreed in writing before bulk begins.

Writing the Inspection Clause Before Bulk Starts

An inspection clause is six lines long and takes an hour to agree: standard, level, AQL per defect class, sample timing, who counts, and what happens on rejection.

The arithmetic of rejection is worth stating plainly. Reworking one closure on a returned piece costs around USD 4.80 in parts and bench time before any freight is counted, so a rejected lot is worth considerably more than a price concession argued for afterwards.

Against that clause, the rest of the programme is unchanged: MOQ 500 pcs per style and colourway, sampling in 6-10 working days and 12-15 for complex constructions, bulk in 35-50 days, and handover on FOB Xiamen terms.

Frequently Asked Questions

People Also Ask