Leather Handbag Grading: What Buyers Notice Before Stitching

What does leather handbag grading actually check before the first stitch is sewn? It counts defects against a threshold and draws samples against a standard, and both are numbers. At 500 pieces, ISO 2859-1 general inspection level II gives code letter J, a sample of 80 units and an accept number of 5 at AQL 2.5; grading itself sorts panels into selection grades A to D, where grade A allows zero to one defect above 3 mm on a visible panel and yields 85-90% of the hide. The difference between a programme that passes first time and one that re-sorts is almost always whether those numbers were written into the specification before cutting began.

Grading numbers to write down

Grading Is a Numbers Exercise, Not an Opinion

Grading is the step that happens before a needle moves, and it is the step most specifications skip. It is also the only step that can be described entirely in numbers: how many marks are allowed, how big they may be, where on the panel they may sit, and how many units an inspector draws to decide whether a lot passes. Once those four numbers exist, grading stops being a discussion about taste and becomes a pass or a fail.

The reason it matters commercially is yield. A grade A selection takes 85-90% of a hide and a grade C selection takes 94-97%, so the same hide priced the same way produces materially different numbers of bags depending on where the grade line is drawn. Most disagreements about leather cost are really disagreements about that line, and neither party notices because neither wrote it down.

• How many marks? Selection grades are defined by defect count, not by appearance.
• How big? A 3 mm mark on a front panel and a 25 mm mark under a lining are not the same defect.
• Where? Zone 1 panels are seen every time the bag is picked up; zone 3 panels are never seen.
• How many units get checked? ISO 2859-1 sets that, and it is not negotiable at the inspection table.

The Five Surface Defects and Their Allowances

Five defect types account for almost all grading decisions on a hide, and each has a different threshold because each behaves differently once the panel is cut, folded and sewn. Scar tissue is stable; a vein mark can open; a loose grain area will crease permanently.

Loose grain is the one defect with no allowance anywhere on a bag, because it is the only one that gets worse in service. A loosened area creases under load, the crease becomes a crack and the crack becomes a return. Every other defect in the table is cosmetic and can be traded for yield; this one cannot. It is also the defect least likely to be caught late, because it looks flat on a bench and only appears when the panel is flexed across the grain.

Zone definitions are the second half of the table, and they are a design decision as much as a quality one. A tote front panel, a flap and the top of a gusset are seen every time the bag is picked up, so they sit in zone 1. Sides and the outer base are seen occasionally, and facings, pocket bags and inner gussets are effectively never seen once the bag is in use. Drawing those three zones onto the pattern before cutting is an hour of work that settles every downstream argument about whether a mark matters.

Counting Defects: What an Inspector Actually Does

An inspector does not look at 500 bags and form a view. Under ISO 2859-1 the inspector looks up the lot size, finds the sample size code letter, draws that many units at random and counts defects by class. At AQL 2.5 the decision is arithmetic: a certain number of major defects passes the lot, one more fails it, and the report records which.

Two things follow from that table, and both surprise first-time buyers. The sample is a fixed count rather than a percentage, so the inspection effort per unit falls sharply as the lot grows; and the accept number at 500 pieces is 5, which means five defective units in a sample of 80 still pass. A programme that wants fewer than that specifies a tighter AQL and pays for it in re-sorting time.

What a Graded Panel Costs at Each Selection Level

Selection level and price move together, and the mechanism is yield rather than quality. A tannery grading to A rejects the parts of a hide that a C selection would cut, so the A panels carry the cost of everything that was set aside.

Splitting the buy across two selection levels is the single largest legal saving in a leather programme. Where a tote front panel needs grade A, the facings and pocket bags almost never do, and straps cut from grade B at 2.0-3.0 mm perform identically once the edge is folded and painted. That split typically removes 8-14% from the material line without any visible change to the bag.

Matching Panels Across a 500-Piece Run

Grading decides whether a single panel is acceptable. Matching decides whether 500 of them look like the same product, and it is a separate specification line with its own numbers. Three properties have to match across a run, and each has a measurable tolerance.

• Colour, matched to a Pantone TCX reference at Delta E 2.0 or better against the approved swatch, checked under D65 and under a store light source such as TL84 to catch metamerism.
• Substance, at plus or minus 0.1 mm measured at three points on three panels, because a 0.2 mm drift changes how a structured panel stands.
• Grain direction, held within 5 degrees of the specified axis on structured silhouettes, since grain running across a gusset creases differently from grain running with it.
• Temper, judged by hand against the retained reference, since it is not measurable on a calliper and is the property customers notice first.

All four are checked on the sample before 500 pieces are cut, and all four belong in the specification with a number attached. A programme that specifies colour by name and substance by feel has no recourse when the shipment differs, because there is nothing to measure it against.

What to Check Before the First Stitch Is Sewn

The checks that catch the most expensive failures all happen before assembly, and they take minutes per hide. Six of them cover the ground, and each maps to a failure that cannot be fixed once the panels are joined.

1. Hold the panel up to the light and look through it for thin spots and loose grain, which are invisible flat on a bench.
2. Flex the panel across the grain at the point where the bag will fold, and watch whether the grain lifts.
3. Measure substance at three points with a calliper against the plus or minus 0.1 mm tolerance.
4. Check the finish for rub with a white cloth against ISO 11640, before the colour transfers onto a lining.
5. Confirm the shrinkage temperature to ISO 3380 where the destination is hot: 85 °C or above for chrome, 75 °C or above for vegetable tannage.
6. Match the lot number on the bundle to the test report, and refuse any bundle that has arrived without one.

Then check the construction set that applies regardless of grade: Tex 40 thread at 7-9 stitches per inch with a 10 mm seam allowance, edge paint in three coats at 0.3-0.5 mm total build, and zinc alloy hardware with plating thickness in microns and 24 hours to first corrosion tested to ASTM B117. Grading catches the material failures; these catch the assembly ones.

Recording the Grade So the Repeat Matches

A grade that exists only in a cutter's head is lost the moment that cutter moves on, and the repeat order starts from a fresh argument. Five records preserve it, and together they are what makes a second order faster and cheaper than the first.

• The signed swatch card, with the grade, substance, finish and colour reference written on it rather than beside it.
• The defect tally sheet for the lot, showing which selection grade each panel set was cut from.
• The lot number and the tannery reference, matched to the test report issued before cutting.
• The retained tooling, identified and stored, so the embossing strike is identical next season.
• The inspection report at AQL 2.5 under ISO 2859-1, including the defect counts by class.

With those five on file, a repeat at an unchanged specification can skip the 6-10 working days of sampling window entirely and run straight to bulk. Without them, the second order is a first order with a longer email thread, and the price does not improve.

What to Put in the Grading Clause

A grading clause is a paragraph in the specification, and it is the paragraph that prevents most disputes. Eight items, each a number or a named method, and none of them longer than a line.

1. Selection grade by panel position: front and flap panels grade A, sides and backs grade B, facings and pocket bags grade C.
2. Defect threshold: nothing above 3 mm in zone 1, nothing above 10 mm in zone 2, nothing above 25 mm in zone 3.
3. Absolute exclusion: no loose grain at any size on any load-bearing panel.
4. Substance and tolerance, in millimetres, plus or minus 0.1 mm.
5. Colour reference and tolerance: Pantone TCX at Delta E 2.0 or better, checked under two light sources.
6. Grain direction, within 5 degrees of the specified axis on structured panels.
7. Test methods: ISO 11640 rub fastness, ISO 3380 shrinkage, ISO 3377-2 tear strength, REACH Annex XVII screening.
8. Inspection standard: ISO 2859-1 general level II at AQL 2.5, with the written report issued before shipment.

Both BSCI and Sedex audit references, together with an ISO 9001 quality system reference, can be supplied for our validated partner factory Quanzhou Yingming Bags, and third-party inspection by SGS can be booked alongside the pre-shipment report. Related detail sits in the full grade buying guide, and the full programme is set out in the leather handbags reference.

Where Zone Boundaries Should Sit on Each Silhouette

A zone map is the most argued-over part of a grading clause, and the argument is usually about a panel nobody sees. Zone one should be the surfaces in a customer's line of sight while the bag is being carried: the front panel, the flap, the top line and the outer face of the gusset. Zone two is seen on handling: the back panel and the sides. Zone three is never seen at all: under a lining, inside a pocket bag, behind a base insert.

• Zone one, seen every time: front panel, flap, top edge and outer gusset face, with a 3 mm defect threshold.
• Zone two, seen on handling: back panel and side panels, with a 10 mm threshold.
• Zone three, never seen: pocket bags, facings, and any area under a lining or a base insert, at 25 mm.
• The rule that settles disputes: a panel only leaves zone three if something is permanently fixed over it on the finished bag.

How Much a Grade A Selection Costs in Hides

Selection level is paid for in hide area rather than in leather quality. Where a tannery grades to A it rejects the parts of the hide carrying marks, and the price per usable square foot rises to cover the discarded area. The arithmetic is worth doing before the grade is written into a specification, because the gap between an all-grade-A buy and a split buy is the largest legal saving available in a leather programme.

The Cutting Allowance That Goes With Each Grade

Grading changes the cutting allowance, and the allowance is what converts a panel count into a hide order. A grade A panel needs large unbroken areas, so the nest wastes more; a grade C panel can be built from smaller pieces, so it wastes less. A programme that specifies grade A across the whole bag pays twice, once for the selection and once for the yield.

• Grade A panels: 18-22 per cent cutting waste, set by the largest unbroken panel on the pattern.
• Grade B panels: 14-18 per cent, since marks can be placed outside the visible area.
• Grade C panels: 10-14 per cent, because small parts can be nested around the defects.
• The saving: reducing the largest unbroken panel by 4 cm typically cuts waste by 2-4 percentage points.

Reading a Tannery Grading Sheet Before You Accept It

Tanneries grade on their own sheets, and the sheet is worth reading before it becomes part of a specification. Four items decide whether it will work as a contract document: the defect types named, the measurement method, the panel the grade applies to, and the lot reference. Where any of the four is missing, the grade is a description rather than a standard, and the scars, scratches and defects guidance is the fallback.

• Defect types named: scar, insect bite, vein mark, growth wrinkle and loose grain each need their own threshold.
• Measurement method: longest dimension in millimetres, measured flat, not judged by eye.
• Panel coverage: a grade stated for a whole hide is not a grade for a front panel.
• Lot reference: a sheet without the lot number cannot be matched to the shipment or the test report.

Frequently Asked Questions

People Also Ask