MOQ from 100 unitsFree design serviceOEM · ODM · Private LabelISO 9001 certified factoryWorldwide DDP shipping18+ years export experience50+ countries served MOQ from 100 unitsFree design serviceOEM · ODM · Private LabelISO 9001 certified factoryWorldwide DDP shipping18+ years export experience50+ countries served
Quality Control

Third-Party Inspection Specs for Custom Pins and Keychains

10 min readBy the ZheCraft team2026-06-13
Third-Party Inspection Specs for Custom Pins and Keychains

Why vague inspection briefs still let bad goods ship

Most quality failures in custom metal giveaways do not happen because nobody inspected the order. They happen because the instruction was too broad: “check quality before shipment.” An inspector then counts cartons, opens a few bags, compares color by eye, and the shipment can still arrive with loose butterfly clutches, scratched plating, weak magnets, mixed SKUs, upside-down backing cards or cartons that exceed courier weight limits.

Pins, brooches, challenge coins, keychains and magnets need an inspection plan that turns the approved sample into measurable checkpoints. The brief should state AQL levels, defect classes, sample size, visual standards, measurement tolerances, functional tests, packing rules and the escalation path after failure. A useful inspection appendix is normally 2 to 4 pages. It does not need legal language; it needs accept/reject limits the inspector can apply on the factory floor.

For China-made promotional metal items, the highest-risk gaps are usually hardware strength, plating consistency, enamel fill, color match, SKU segregation and retail packing. At ZheCraft, in-process QC is built into the Yiwu production workflow, but many international buyers add a third-party pre-shipment inspection for licensed programs, retail launches, museum shops, distributor orders and anything with strict receiving rules. The goal is not to reject handmade variation. The goal is to catch defects that create returns, complaints, safety issues or chargebacks before cartons leave China.

Schedule inspection when it can still change the outcome

Final random inspection should be booked when 100% of production is finished and at least 80% is packed. For small metal items, inspecting at 100% finished but only 40% to 50% packed is too early because many failures occur during packing: wrong polybag, missing warning label, mixed artwork revision, short inner box count, skewed backing card, incorrect carton mark or overweight export carton. The inspector needs access to finished loose goods and enough sealed cartons to verify the real shipping configuration.

For orders above 20,000 pieces, or any order with more than five SKUs, add an in-line inspection when 20% to 30% of production is complete. This is the stage to catch systemic problems before they are repeated across thousands of units: plating thickness below specification, enamel overflow from a shared mold, die-struck recesses too shallow, weak spring temper in keyrings, poor magnet bonding or incorrect backing-card artwork. In-line inspection does not replace final inspection; it reduces the cost of failure.

Build time into the logistics plan. Schedule final inspection 3 to 5 working days before vessel closing, courier pickup or air cargo handover. Simple sorting of 5,000 pins may take 1 to 2 days. Replacing clutches, repacking retail cards or relabeling cartons may take 2 to 4 days. Replating, remaking enamel, replacing welded posts or producing new hardware normally takes 5 to 12 days, depending on plating line capacity and mold availability. If inspection is booked on pickup day, the buyer often has only two choices: ship known defects or miss the launch.

Order profileRecommended inspection pointReadiness requirementTypical inspection time
300 to 999 pcs, 1 to 3 SKUsFinal random inspection100% finished, at least 80% packed0.5 day
1,000 to 4,999 pcsFinal random inspection100% finished, 80% packed, carton marks ready0.5 to 1 day
5,000 to 19,999 pcsFinal inspection plus optional line checkLine check at 20% to 30%; final at 80% packed1 to 2 days total
20,000 pcs and aboveIn-line inspection plus final random inspectionLine check before full plating or assembly; final at 80% packed2 to 3 days total
Multi-SKU set or retail packFinal inspection after set assembly100% components finished; 80% complete sets packed1 to 2 days
Licensed or premium retail orderPilot run, in-line check and final inspectionGolden sample approved; first 500 to 1,000 pcs reviewed before scale-up2 to 4 days total

Set AQL levels and sample sizes that match the risk

For standard promotional pins and keychains, a balanced starting point is AQL 0 for critical defects, AQL 1.5 for major defects and AQL 4.0 for minor defects under ANSI/ASQ Z1.4 or ISO 2859-1, normal inspection, general inspection level II. For premium retail, licensed characters, museum merchandise or brand campaigns with high complaint exposure, tighten to AQL 0 critical, AQL 1.0 major and AQL 2.5 minor. For low-cost event giveaways, AQL 2.5 major and AQL 4.0 minor may be acceptable if the buyer accepts the higher field-complaint risk.

Avoid writing “zero defects” for every issue. Custom enamel and plated metal parts are hand-finished. A 0.1 mm enamel meniscus against a raised metal line, a dust dot under epoxy smaller than 0.3 mm outside the logo, or a short polishing direction mark on the back may be normal if it is not visible at ordinary viewing distance. Unrealistic cosmetic standards cause shipment holds without improving end-user experience.

Critical defects remain zero tolerance. These include sharp burrs that can cut skin, exposed needle points, detached pin posts, magnet ingestion risk on child-related items, wrong logo, wrong customer design, missing legal warning label, rust, mold mismatch against the approved sample or mixed customer property. One critical defect in the inspection sample should trigger shipment hold, root-cause review and 100% sorting of the affected SKU or process.

The sample pull should be random and traceable. Under general inspection level II, a lot of 3,200 pieces commonly produces a sample size of 200 pieces; a lot of 10,000 pieces commonly produces 315 pieces; a lot of 35,000 pieces commonly produces 500 pieces. For mixed SKUs, the inspector should pull from beginning, middle and end cartons, not from cartons presented by the packing team. If one SKU is safety-sensitive or retail-critical, assign a separate sample to that SKU instead of hiding it inside the combined lot.

  • State the standard: ANSI/ASQ Z1.4 or ISO 2859-1, normal inspection, general level II.
  • Use AQL 0 critical, 1.5 major and 4.0 minor for most B2B promotional orders.
  • Tighten to AQL 0 critical, 1.0 major and 2.5 minor for retail, licensed or premium orders.
  • Define separate sampling for high-risk SKUs, child-adjacent products and magnetic attachments.
  • Confirm who pays sorting and reinspection when failure is caused by factory workmanship.
  • Require written buyer approval before shipment after any critical defect, even if the count is low.

Classify defects with measurable visual standards

The inspection brief should separate defects into critical, major and minor. A major defect affects saleability, function, brand appearance or normal use even if it is not dangerous. Examples include loose clutches, a split-ring gap above 0.5 mm, enamel missing in a visible area, plating stain larger than 1.0 mm outside the logo, Pantone mismatch beyond the approved sample range, an unreadable QR code or a backing card printed with an old revision.

Minor defects are limited cosmetic issues a normal customer is unlikely to notice during ordinary handling. For pins and coins, this may include a polishing hairline under 3 mm outside the focal area, a plating speck under 0.3 mm away from text or face detail, or a slight enamel surface variation that does not expose base metal. Minor defects still matter because a high concentration makes the order look cheap and can fail retail receiving standards.

Specify how appearance is judged. A practical standard is visual inspection at 30 to 40 cm under 600 to 1,000 lux neutral white light, without magnification, for 5 to 8 seconds per side. Use magnification only for features that require it: QR readability, serial numbers, line width, recessed text, plating thickness test points or tooling details on a technical drawing. Color should be checked against the approved physical sample or Pantone chip under D65 or neutral daylight, not yellow warehouse lighting.

DefectClassReject threshold
Sharp burr, exposed needle point, detached pin postCriticalAny occurrence
Wrong artwork, mixed customer design, missing legal warningCriticalAny occurrence
Loose clutch, failed brooch bar, weak magnet bondMajorAny occurrence in sampled piece
Enamel missing, overflow into logo line, wrong filled areaMajorVisible at 30 to 40 cm
Plating blister, peeling, rust or black spot on logoMajorOver 0.5 mm in logo area or over 1.0 mm elsewhere
Pantone mismatch against approved sampleMajorClearly visible under neutral light or outside agreed Delta E limit
Polish hairline outside focal areaMinorVisible but under 3 mm and not clustered
Dust dot under epoxyMinorUnder 0.3 mm and outside logo, face or text area
Backing card skew or bag wrinkleMinorUnder 3 mm if barcode and brand view remain acceptable

Measure specifications that drive returns

Many inspectors can count cartons and photograph defects, but metal-product measurement is inconsistent unless the buyer defines tools and tolerances. Require a digital caliper with 0.01 mm resolution and measurement of 5 to 13 pieces per SKU, depending on order value and risk. For stamped zinc alloy, brass or iron pins, a practical tolerance is ±0.20 mm for overall length or width under 50 mm, ±0.30 mm for 50 to 80 mm and ±0.50 mm for larger cast coins, bottle openers or magnets. Hole diameter, post position and magnet position should be tighter where assembly depends on fit.

Measure base-metal thickness separately from raised relief and epoxy dome. A 1.5 mm soft enamel pin may be acceptable at 1.45 to 1.65 mm on the base metal. A domed epoxy keychain may read 2.0 to 2.8 mm at the highest point, but that number should not be treated as metal thickness. Weight tolerance is normally ±5% for die-struck or cast metal items and ±8% for complex 3D relief, spinner parts or mixed-material keychains.

Plating thickness should be stated in microns only when it will be tested by XRF or supported by plating-line records. Standard promotional flash plating may be only 0.03 to 0.05 micron for the visible color finish, which is low-cost but not highly wear-resistant. A more durable specification may call for a 0.10 to 0.20 micron nickel underlayer plus a 0.05 to 0.10 micron gold, silver, black nickel or antique finish where compatible. Premium coins, bottle openers and frequently handled keychains may require thicker plating, but every extra micron increases cost and lead time.

FOB China pricing varies sharply by size, metal, plating, packing and order quantity. As a rough 2025 planning range, 25 to 35 mm soft enamel pins are often USD 0.45 to 0.90 FOB at 1,000 pieces and USD 0.25 to 0.55 at 10,000 pieces. Zinc alloy keychains are often USD 0.70 to 1.60 at 1,000 pieces and USD 0.45 to 1.10 at 10,000 pieces. Challenge coins may run USD 1.20 to 3.80 depending on diameter, thickness, 3D relief and edge finish. These ranges are not substitutes for quotation, but they help buyers see when an unrealistic price forces weak plating, thin metal or rushed packing.

  • Overall size: record length and width against the production drawing, not only the artwork proof.
  • Base thickness: measure away from raised logo, welded post, relief peaks and epoxy dome.
  • Plating thickness: use XRF if the contract states microns; visual inspection cannot verify thickness.
  • Attachment position: check pin posts, magnets and keychain loops against tolerance because off-center hardware causes returns.
  • Color: compare to an approved physical sample or Pantone chip under D65 or neutral white light.
  • Packing count: reconcile pieces per bag, inner box and export carton with carton marks and packing list.

Run functional tests on hardware, magnets and finishes

A strong pre-shipment inspection includes simple functional tests that take minutes on the inspection table. For butterfly clutches, attach and remove the clutch 5 cycles, then confirm it still grips the post under light hand pull. For rubber clutches, inspect hole centering, oiliness and grip after 10 push-pull cycles. A pin post should remain straight after manual pressure and should not rotate or detach from the base.

For keychains, open and close lobster clasps or carabiners 10 cycles and confirm the gate returns flush every time. Split rings should not show permanent opening above 0.3 mm after inserting the actual charm loop or a 1.5 mm test tab. Jump rings should be fully closed, with a visible gap below 0.2 mm for standard promotional items. For charms above 30 g, welded rings, double-loop rings or heavier-gauge split rings reduce field failure.

For magnets and magnetic badges, test the intended use, not only magnetic attraction by hand. A fridge magnet should hold its own weight plus one A4 sheet on a painted steel panel for 30 seconds. A magnetic badge should not slide on a 180 to 220 gsm cotton shirt during a 30-second shake test. Magnetic attachments should not be used for child-focused products, pacemaker-sensitive environments, very thin fabrics or security badges where loss creates operational risk.

Product or featureInspection testSuggested pass standard
Pin with butterfly clutch5 attach-remove cycles plus light hand pullClutch does not slide; post stays straight
Rubber clutch10 push-pull cyclesGrip remains firm; no tearing or oily residue
Brooch barOpen-close 10 cyclesLatch closes fully; hinge is not loose
Split ringInsert actual loop or 1.5 mm tabNo permanent gap over 0.3 mm
Lobster clasp or carabinerOpen-close 10 cyclesSpring returns gate flush each cycle
Fridge magnetHold own weight plus one A4 sheet on painted steelNo sliding for 30 seconds
Epoxy domeThumb pressure at dome edgeNo lifting, clouding, cracking or adhesive failure
Challenge coin finishRub with dry white cotton cloth 20 strokesNo visible plating transfer or color smear

Control MOQ, lead time, packing and SKU mix

Inspection should also verify commercial assumptions that affect delivery. Typical factory MOQs are 100 to 300 pieces per design for simple enamel pins, 300 to 500 pieces for keychains or magnets with custom molds and 500 to 1,000 pieces for retail-packed sets where printed cards, inserts or barcodes are customized. Small MOQs are possible, but unit price rises because mold charge, plating setup, printing setup and packing labor are spread across fewer pieces.

Normal production lead time after artwork and sample approval is usually 12 to 18 days for simple enamel pins, 15 to 25 days for zinc alloy keychains, 18 to 30 days for challenge coins with 3D relief and 20 to 35 days for multi-component retail packs. Add 3 to 7 days for physical sample making before mass production, and add 5 to 10 days if special plating, epoxy curing, printed inserts or licensed packaging approvals are required. Inspection should be booked only after these process steps are complete, not while cards or cartons are still arriving.

Packing is a frequent failure point. For orders with more than three designs, require SKU segregation at every layer: unit bag if needed, inner box, export carton and master packing list. Carton marks should show PO number, SKU code, design revision, quantity, carton number, gross weight and destination. For courier networks, keep export cartons below 15 kg for pins and keychains and below 12 kg for heavy coins unless the forwarder agrees otherwise. Common small-metal cartons are 35 x 25 x 25 cm or 40 x 30 x 25 cm in 5-ply corrugated board; heavy coins may need double-wall cartons, smaller pack counts and reinforced tape.

Drop checks should be agreed before inspection because they can damage saleable goods. A practical non-lab check is 1 corner, 3 edges and 6 faces from 60 cm for cartons under 10 kg, followed by review of inner boxes and product condition. For retail packaging, passing means more than product survival. Blister cards, backing cards, gift boxes and printed sleeves must still look shelf-ready.

  • Open cartons from the beginning, middle and end of the packed lot, not only factory-selected cartons.
  • Verify mixed-set assembly by counting every component inside at least 32 sets or the AQL sample size, whichever is larger.
  • Scan barcodes and QR codes on backing cards and retail boxes with a normal phone or scanner.
  • Photograph one approved packed-carton configuration before sealing the shipment.
  • Record actual carton dimensions and gross weight for freight invoice and warehouse receiving.
  • Check that carton quantity, PO quantity and packing list quantity reconcile by SKU.

Use inspection as a decision tool, not a cure-all

Third-party inspection is a sampling tool. It estimates lot quality, but it does not guarantee every piece is defect-free unless the buyer pays for 100% sorting. If the product uses translucent enamel over textured plating, moving spinner parts, large epoxy domes, thin cutouts, magnetic attachments, soft PVC bonded to metal or child-adjacent packaging, the buyer should approve process samples and stress tests before mass production.

Inspection also cannot repair unclear artwork. If the approved proof contains a 0.15 mm raised metal line, an enamel island under 0.4 mm, a magnet too close to the edge or text too small to fill cleanly, the inspector can only judge the delivered goods against that weak design. The better fix is manufacturability review before tooling: increase minimum line width, enlarge enamel islands, adjust magnet diameter, add a second pin post, change plating or simplify cut-through details.

Before placing the purchase order, attach an inspection appendix to the RFQ or proforma invoice. List the approved sample reference, product drawing, AQL levels, defect definitions, tolerances, functional tests, packing specification, inspection photos required and the rule for reinspection after failure. If a supplier rejects reasonable inspection terms before production, treat that as a sourcing warning.

For a normal 1,000 to 10,000 piece pin or keychain order, a practical baseline is AQL 0 critical, 1.5 major and 4.0 minor; general inspection level II; 30 to 40 cm visual distance; 600 to 1,000 lux light; ±0.20 to ±0.30 mm dimensional tolerance for most small parts; and functional testing on 13 to 32 pieces per hardware type. For retail, licensed or child-adjacent items, tighten major AQL, add safety-specific checks and require traceability by SKU and carton. Sending this standard before sampling gives the factory, ZheCraft QC team and third-party inspector one objective target, which reduces shipment holds, rework disputes and surprise claims after the cartons have left China.

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

Ready to get this made?

Send your sketch, target quantity and ship-date. Detailed quotation in 12 hours.

Start Your Project »