Quality Control

AQL Inspection Plans for Custom Metal Giveaways

Why Good Samples Still Become Bad Shipments

A polished pre-production sample does not guarantee that 5,000 bulk pieces will match it. Custom metal giveaways fail in bulk for predictable reasons: plating shade drift between racks, enamel dust under epoxy, weak jump rings, loose pin posts, mixed backing cards, short counts, or magnets that detach after packing. These problems usually come from an incomplete inspection plan, not from a factory deliberately ignoring quality.

For enamel pins, brooches, keychains, fridge magnets and challenge coins, inspection must separate appearance, function, dimensions, plating and packing. A 25 mm soft enamel pin on a card does not need the same pull test, carton weight limit or cosmetic threshold as a 50 mm die-struck challenge coin in a capsule. Buyers reduce disputes when the purchase order states how many units will be checked, what counts as a defect and which results stop shipment.

This guide is for B2B orders from about 300 to 50,000 pieces per SKU. It is not a medical or aerospace standard. It is a practical factory-floor QC plan for promotional metal products where the goal is consistent brand appearance, safe handling and low return risk without adding unnecessary inspection cost.

Classify Defects Before Choosing AQL

Start with defect classes. A critical defect is unsafe, illegal or likely to create a serious claim. Examples include a sharp burr above 0.15 mm on a wearable brooch, a broken pin post, a detached small magnet, a choking-risk component on a child-facing item, or a product marked nickel-free without supporting material control. Critical defects should have an acceptance number of 0 in the inspected sample.

A major defect prevents normal use or damages brand value. Typical major defects include wrong Pantone color outside the approved sample range, missing logo elements, plating blisters larger than 1 mm, enamel overflow covering metal lines, epoxy scratches visible at 30 cm, a key ring that opens under the agreed pull force, or a backing card printed with the wrong QR code. These thresholds must be written clearly because most buyer-factory disputes happen in the major category.

A minor defect is visible on close inspection but does not affect normal use. Examples include a back-side hairline scratch under 3 mm, a dust point under epoxy below 0.3 mm, a slightly skewed polybag label, or small shade variation within the signed golden sample range. Minor defects should still be tracked, but rejecting a complete shipment for a few small reverse-side marks usually creates more cost than value.

Defect class	Typical AQL	Concrete examples	Shipment action
Critical	0.0	Sharp burrs over 0.15 mm, broken attachments, unsafe magnets, false safety claim	Hold shipment; sort or remake affected goods
Major	1.5 or 2.5	Wrong color, loose hardware, plating blister over 1 mm, missing print, wrong card	Reject or rework if over limit
Minor	4.0	Tiny back scratches, dust points under 0.3 mm, slight label skew	Accept within limit; correct next run
Measurement	By tolerance	Size, thickness, plating microns, magnet pull, pull force	Judge against numeric specification

Set AQL Levels and Sample Sizes by Risk

For most corporate promotional orders, a practical default is AQL 0.0 critical, 2.5 major and 4.0 minor under ANSI/ASQ Z1.4 or ISO 2859-1, general inspection level II. For licensed characters, retail programs, VIP gifts or high-visibility launches, tighten to AQL 1.5 major and 2.5 minor. For short-use event giveaways, AQL 2.5 major and 4.0 minor is usually workable if critical defects remain at 0.

Sample size must be stated with the inspection level. Under general level II, a 1,000-piece order commonly gives an 80-piece sample; 5,000 pieces commonly gives 200 pieces; and 20,000 pieces commonly gives 315 pieces. The accept and reject numbers then come from the AQL table. For example, at AQL 2.5 with an 80-piece sample, the shipment normally accepts up to 5 major defects and rejects at 6. At a 200-piece sample, it accepts up to 10 and rejects at 11.

Do not tighten AQL without budgeting for sorting. AQL 1.0 major on a 10,000-piece low-cost keychain order may force reinspection or 100 percent sorting, adding 1 to 3 working days and about USD 0.02 to 0.06 per piece depending on hardware complexity. If the FOB price is only USD 0.55 to 1.20 per zinc alloy keychain, excessive inspection can consume the budget that should have been used for better plating, stronger rings or thicker bags.

• Use AQL 0.0 for safety, legal claims and mixed-SKU errors that create brand risk.
• Use AQL 1.5 major for retail, licensed artwork, VIP gifts and launch events.
• Use AQL 2.5 major for standard corporate promotions and distributor stock orders.
• Use AQL 4.0 minor when small cosmetic marks on non-display surfaces are acceptable.
• Specify general inspection level II unless the order is low-risk enough for level I or risky enough for tightened inspection.

Put Numeric Tolerances in the QC Sheet

AQL defines how many defects are allowed; it does not define what a defect is. The QC sheet must state measurable tolerances for size, thickness, plating, color, weight and attachment position. Without numbers, one inspector may accept a batch while another rejects the same cartons.

For die-struck iron, brass or zinc alloy pins under 40 mm, use outer dimension tolerance of ±0.3 mm. For 40 to 80 mm coins, badges and keychains, ±0.5 mm is more realistic. Thickness tolerance is usually ±0.2 mm for 1.2 to 2.0 mm badges and ±0.3 mm for 3.0 to 5.0 mm challenge coins or bottle-opener keychains. Pin post or magnet position should normally be within ±1.0 mm; tighten to ±0.5 mm only when balance, rotation or a fitted tray requires it.

Plating should be specified by thickness and finish, not only by color name. Standard nickel, gold, black nickel or antique finishes for promotional pins are commonly 3 to 5 microns total deposit thickness. Heavier keychains, coins and items that rub against keys should target 5 to 8 microns. Plating below about 2 microns may pass a quick visual inspection but is more likely to expose copper or base metal on edges after handling.

Specification point	Practical target	Tighten when
Outer size under 40 mm	±0.3 mm	Backing card windows, paired sets, fitted boxes
Outer size 40 to 80 mm	±0.5 mm	Coin capsules, molded trays, paired magnets
Thickness	±0.2 to ±0.3 mm	Spinner coins, closures, retail displays
Plating thickness	3–5 microns standard; 5–8 microns heavy use	Keychains, coins, high-rub applications
Hard enamel flatness	0.05–0.10 mm enamel-to-metal height difference	Premium retail pins, jewelry-style badges
Pantone color	Approved sample; Delta E 2.0–3.0 if measured	Licensed colors, brand-critical programs
Attachment position	±1.0 mm standard; ±0.5 mm functional	Large brooches, anti-rotation pins, centered hangers

Inspect Function, Not Only Appearance

Many metal giveaways look acceptable on the table and fail during use. For butterfly clutch pins, inspectors should attach and remove the clutch 3 to 5 times and confirm that it still grips the post. For rubber clutches, check tearing around the hole and whether the clutch slides off under light finger pressure. A pin with a clean enamel face but a weak clutch becomes a complaint as soon as staff start wearing it.

For keychains, test the split ring, jump ring, chain and connector as one system. A practical pull test is 5 kg for light charms and 8 to 10 kg for heavier zinc alloy keychains or bottle-opener styles, held for 10 seconds. A quick hand tug is not a test. If a connector gap opens, the ring deforms or the chain separates, classify the unit as major because the item can be lost in normal use.

For fridge magnets and magnetic badges, specify both magnet type and holding force. A 20 mm ferrite magnet may hold a lightweight PVC piece but underperform on a 45 mm die-cast badge. For a 35 to 45 mm metal fridge magnet, a practical target is 250 to 500 g pull force on a clean steel plate, depending on total item weight. If the magnet is glued, require at least 24 hours curing before final pull testing; fresh adhesive can give misleading results.

• Check at least 20 pieces per SKU for clutch fit, key ring closure or magnet adhesion during final inspection.
• Use 5 kg pull for light keychains and 8–10 kg for heavier metal keychains unless the design requires more.
• Reject pin posts that rotate, wobble or detach under moderate thumb pressure.
• Reject magnets that detach from the product before the magnet itself loses hold.
• Record failures by component so rework targets the post, ring, chain, adhesive or magnet.

Control Color, Plating and Finish With Samples

Color standards need a hierarchy. The approved pre-production sample should control overall appearance, the Pantone number should control intended enamel or print color, and the production artwork should control layout. If these conflict, the QC sheet must say which one wins. Transparent enamel over gold plating will not look like the same Pantone printed on white paper, so visual approval against a real sample matters more than a screen image.

Plating shade variation is normal between batches, especially antique gold, antique silver, rose gold and black nickel. Instead of demanding impossible perfect sameness, approve a signed golden sample. For larger programs, approve upper and lower shade limits. Inspect polished gold and nickel under neutral 5000K to 6500K light; do not judge plating shade under yellow warehouse lamps. A finish that looks too warm at 3000K may match the sample in daylight.

Surface finish rules must match the process. Soft enamel pins normally have recessed enamel and raised metal lines. Hard enamel should be polished nearly flat, with enamel-to-metal height difference usually within 0.05 to 0.10 mm. Epoxy-coated pieces should have clear doming without bubbles, orange peel, trapped fibers or edge overflow. Antique coins should have dark recessed areas, but black residue on raised faces, text or portrait details is a major defect if it hides the design.

Check Packing Before Cartons Close

Packing errors are expensive because they often reach the final recipient before anyone notices. A shipment can have acceptable products but still fail if 30 mm pins are packed on the wrong card, keychains are mixed between languages, QR codes point to the wrong campaign, or carton labels show the wrong event date. Inspection must cover retail units, inner cartons and master cartons.

For polybagged pins and keychains, specify one product per bag, correct clutch or ring included, no trapped hardware marks on the display face, and bag size large enough to avoid bending the card. A 30 mm pin on a 55 x 85 mm backing card commonly uses an OPP bag around 70 x 100 mm. For coins and heavier keychains, use 0.04 to 0.06 mm bags instead of 0.025 mm bags, which tear more easily during carton vibration.

Master cartons should usually stay under 15 kg gross weight for courier shipments and under 18 kg for palletized sea freight unless the buyer approves otherwise. Small pins are often packed 500 to 1,000 pieces per carton; challenge coins are more practical at 100 to 300 pieces per carton depending on diameter, capsule, pouch or box. Oversized cartons may reduce carton count but increase crushing, drop damage and short-count risk.

Packing checkpoint	Inspection method	Reject condition
Backing card match	Compare SKU, barcode, language and artwork against PO	Correct item on wrong card
Retail bag	Visual check sampled units	Torn bag, missing clutch, scratched front face
Inner count	Count selected inner cartons	Short count or mixed designs
Carton weight	Check gross weight and carton grade	Bulging carton over approved limit
Carton label	Compare marks to shipping instruction	Wrong PO, destination, event name or SKU

Match Inspection Cost to MOQ, Price and Lead Time

The inspection plan should fit the commercial value of the order. A 500-piece rush order of 25 mm soft enamel pins at FOB USD 0.55 to 0.95 per piece cannot carry the same QC burden as a 20,000-piece retail keychain program at FOB USD 1.20 to 2.80 per piece. For small orders, prioritize safety, sample match, function and packing accuracy. For large programs, add in-process checks because final inspection alone may find defects too late.

Typical MOQ is 100 to 300 pieces for simple pins, keychains and magnets, but pricing improves at 500, 1,000 and 3,000 pieces because tooling, plating setup and QC time are spread across more units. Challenge coins and bottle-opener keychains may still be possible at 100 to 300 pieces, but unit prices are more stable from 500 pieces upward. Common FOB ranges are USD 0.45 to 1.10 for simple soft enamel pins, USD 0.70 to 1.80 for metal keychains, USD 0.80 to 2.50 for fridge magnets with metal bodies, and USD 1.20 to 4.50 for challenge coins depending on diameter, plating, enamel and packaging.

Lead time must include inspection and possible correction. Standard pins and keychains often need 5 to 7 days for artwork confirmation and tooling, 5 to 8 days for sampling, and 10 to 18 days for mass production after approval. Challenge coins, 3D badges or multi-item gift sets may need 18 to 30 days in production. Final inspection normally takes 0.5 to 1 day for small single-SKU orders and 1 to 2 days for larger or multi-SKU orders. Reserve another 1 to 3 working days for sorting, repacking or replacement parts if the inspection fails.

Add the QC clause before quotation, not after goods are packed. A concise clause can state: final inspection under ISO 2859-1, general inspection level II, AQL 0.0 critical, 2.5 major, 4.0 minor; outer size ±0.3 mm under 40 mm; plating 3 to 5 microns; keychain pull test 8 kg for 10 seconds; carton gross weight under 15 kg; approved sample controls color and finish. For retail or licensed work, tighten major defects to AQL 1.5 and attach signed golden sample photos.

• Attach approved artwork, Pantone list, packing layout and golden sample photos to the purchase order.
• Define critical, major and minor defects before mass production starts.
• Confirm whether inspection happens at 80 percent packed, 100 percent packed or before retail packing.
• Reserve 1 to 3 working days for inspection, sorting and repacking.
• Request photos of rejected examples so the next reorder improves at the source.

Have a project? Send your artwork and target quantity and we’ll reply with a detailed quotation within 12 working hours.