Quality Control

The 2026 QC Handoff Plan for Custom Metal Promo Orders

Start with handoff risk, not artwork approval

Most custom metal promo orders do not fail because the vector logo was wrong. They fail because quality gets reinterpreted at every handoff: RFQ, engineering review, tooling, sample approval, mass production, packing and final inspection. A buyer may approve artwork, but the factory still has to decide minimum line width, recessed fill depth, plating appearance, hardware location, attachment strength and pack-out accuracy. If those points are not frozen in measurable language, each department applies its own internal standard and drift becomes likely.

For B2B buyers, a usable RFQ needs acceptance criteria by product family, not just size, plating, quantity and logo. Example: a 35 mm soft enamel pin in zinc alloy, nominal thickness 1.5 mm, antique nickel finish, butterfly clutch and individual polybag should also state minimum raised metal line width 0.25 mm, minimum recessed text stroke 0.20 mm, minimum text height 1.2 mm, enamel recession 0.05-0.10 mm below metal line, finished size tolerance +/-0.20 mm, thickness tolerance +/-0.10 mm, hardware position tolerance +/-0.50 mm, burr height less than 0.10 mm after polishing, and inspection plan ANSI/ASQ Z1.4 General Level II, AQL 2.5 major and 4.0 minor.

The most stable programs assign one QC owner before the first quote is requested. That person may sit in sourcing, product development, licensing or with a third-party inspection firm, but one owner must decide what is acceptable and what is not. Without that authority, one team approves a sample from phone photos, another rejects the bulk lot under direct LED light, and the supplier is left guessing which standard controls the order.

1) Build the RFQ so it can become the inspection sheet

The best RFQs already read like page one of a final inspection report. Include overall size in millimetres, nominal thickness, target weight where relevant, base material, process route, finish, enamel type, attachment hardware, packaging method, carton requirement and any compliance limits such as lead, cadmium and nickel release restrictions for destination markets. For stamped pins and badges, finished size tolerance of +/-0.20 mm and thickness tolerance of +/-0.10 mm are practical defaults. For larger die-cast zinc alloy keychains with irregular contours, +/-0.30 mm on overall size and +/-0.15 mm on thickness are more realistic. Color should be specified by Pantone Solid Coated code unless the item is offset printed or CMYK digital printed.

Split lead time by stage instead of using one vague total. For standard custom metal promo products, a practical planning range is 2-3 working days for quotation and art check, 3-5 days for digital proof plus mold layout approval, 7-10 days for pre-production sample, and 12-20 working days for bulk production after sample sign-off. Add 3-7 days for custom backing cards, barcode labels, mixed-SKU kitting, retail blister packs or gift-box assembly. Rush schedules under 15 total production days are possible on simple 2D pins, but the tradeoff is usually higher count error, plating shade variation and weaker sorting discipline.

Commercial terms belong in the RFQ because they affect QC decisions. Ask for MOQ tiers and USD FOB pricing at 100, 300, 500 and 1000 pieces, plus mold or setup charge and any remake policy for defective goods. Typical 2026 FOB ranges are USD 0.42-0.90 for a 30-40 mm soft enamel pin at 500 pcs, USD 0.78-1.75 for a 45-60 mm zinc alloy keychain at 500 pcs, USD 1.15-2.60 for a 45-55 mm challenge coin at 500 pcs, and USD 0.58-1.55 for a 50-80 mm metal magnet at 500 pcs. Adders commonly run USD 0.03-0.08 for epoxy dome, USD 0.05-0.15 for laser serial number, USD 0.04-0.12 for custom backing card insertion, and USD 0.20-0.80 for velvet or rigid gift box depending on size.

• State size, thickness and any critical dimensions in mm, not inches only
• Name the exact process: stamped iron, die-cast zinc alloy, photo-etched brass, die-struck brass or printed stainless steel
• List Pantone codes, finish type and whether slight lot-to-lot shade variation is acceptable
• Define minimum line width and text size, such as 0.25 mm lines and 1.2 mm text height
• Set tolerances for hardware placement: typically +/-0.50 mm for pins and +/-1.00 mm for larger badges or brooch bars
• Specify inspection plan: ANSI/ASQ Z1.4 level, AQL and defect categories
• Call out packaging: polybag, backing card, inner pack count, carton gross weight limit and shipping marks

2) Review quotes for process fit, subcontract risk and hidden exceptions

When quotes arrive, the first job is not comparing unit price. It is confirming whether the proposed process can reliably hold the design. Text below 0.30 mm stroke width is risky in standard stamped iron but reproduces more consistently in photo-etched brass. A large openwork keychain with integrated swivel and welded ring is usually better in die-cast zinc alloy than in stamping with secondary welding. A high-polish 3D coin with deep relief generally needs die-struck brass or a slower-finished zinc-alloy route; otherwise polishing can wash out facial detail, edge steps and small lettering.

Ask which operations are done in-house and which are outsourced. In this category, plating, epoxy doming, offset printing, backing card printing, magnet lamination and retail packing are often subcontracted. Subcontracting itself is not the issue. The risk is weak incoming inspection and poor traceability. If plating is outsourced, ask whether the factory records plating batch date, vendor code and incoming results for color family, adhesion, sidewall coverage, blistering and exposed base metal before assembly starts. For printed backing cards, ask whether they verify barcode scan grade and revision number before use.

Request a written exception list with the quote. This forces suppliers to state what the price does not cover in quality terms. Typical exceptions include slight antique shade variation between lots, minor sink on large die-cast back surfaces, backside swirl marks from polishing, 1-3% overrun or underrun by count, and hairline marks visible only under direct angled light on mirror-finish fields. It is better to negotiate those points before PO release than during a claim after shipment.

Product type	Recommended process	Typical MOQ	Sample lead time	Bulk lead time	Typical FOB range
30-40 mm enamel pin	Stamped iron for flat 2D shapes; die-cast zinc alloy for cut-outs, thicker relief or complex back bosses	100 pcs	7-9 days	12-18 days	USD 0.42-0.90 at 500 pcs
45-60 mm keychain	Die-cast zinc alloy with split ring or swivel; stamped steel only for simple flat tags	100 pcs	7-10 days	15-20 days	USD 0.78-1.75 at 500 pcs
45-55 mm challenge coin	Die-struck brass for sharper 2D relief; zinc alloy for lower-cost 3D forms	100 pcs	8-10 days	15-22 days	USD 1.15-2.60 at 500 pcs
50-80 mm magnet	Stamped or die-cast metal face with ferrite strip or flexible rubber magnet bonded to back	100 pcs	6-9 days	12-18 days	USD 0.58-1.55 at 500 pcs

3) Freeze the pre-production sample with measurable defect language

Sample approval should lock measurable points, not comments like looks good or make colors brighter. Freeze overall size, thickness, weight range if relevant, hardware position, front and back finish, plating family, enamel depth, edge quality, adhesive method, packaging layout and carton marking. If the sample is not treated as the production master, the factory will create its own standard during the run.

Defect definitions must be short enough for line inspectors to use. Example standards: exposed base metal on any front-facing decorative area is major; plating blister, peel or bare sidewall over 1.0 mm is major; dent or pit over 0.30 mm on the front is major; scratch longer than 3.0 mm visible at 30 cm under 800-1000 lux lighting is major on mirror-polished coins and minor on standard pin backs; enamel pinholes under 0.15 mm may be minor if isolated, but three or more within a 10 mm area become major; soft enamel recession up to 0.08 mm below metal line is acceptable, while imitation hard enamel should be nearly level, typically within 0.02-0.03 mm after polishing.

Require a sample record with actual measurements, not just photos. At minimum, record overall dimensions measured at three points, thickness at two points, weight in grams, base material, plating specification, enamel type, attachment test result and packaging sample. Decorative plating in this industry is usually specified by guide range rather than engineering-grade deposition, but declared targets are still useful: 0.03-0.05 microns flash gold, 0.05-0.10 microns decorative nickel, and clear protective lacquer over antique silver when tarnish resistance is required. If a supplier cannot declare a target range, finish consistency will be hard to control.

For adhesive assemblies, define functional tests at sample stage. A fridge magnet should specify magnet material, dimensions and bond method, such as one 20 x 8 x 1.5 mm ferrite strip with pressure-sensitive adhesive plus edge glue, or a full-back flexible rubber magnet sheet 0.7-1.0 mm thick. Require minimum 24 hours cure before bond testing. For pins, define pin-post twist resistance and clutch fit. For keychains, define split-ring wire diameter, for example 1.8-2.0 mm, and require the ring to close with no visible gap above 0.30 mm.

4) Turn the golden sample into a line-ready QC packet

A golden sample only helps if production, QC and packing teams can identify it quickly. Convert the approved sample into a one-page QC packet with front and back photos, ruler reference, Pantone list, hardware photo, packaging photo, approved exceptions and the top five unacceptable defects. This document should travel with the PO, work order and packing instruction. On mixed-SKU orders, issue one packet per SKU. Similar badge sizes, backing cards and plating colors are more often mixed at assembly or final packing than at tooling.

Add an in-process hold point. A practical rule is first-article approval after 30-50 completed units per SKU, before the full lot is finished. For orders above 5,000 pcs per SKU, add one more checkpoint at roughly 10-15% of completed output. These hold points catch the issues that usually appear after startup: wrong backing card revision, antique wash too dark, epoxy with trapped dust, excess glue around magnet strips, keychain swivel assembled backwards, or pin-post placement causing the badge to sit crooked on fabric.

Define lot inspection logic in advance. A common setup is ANSI/ASQ Z1.4 General Inspection Level II with AQL 2.5 for major and 4.0 for minor. Licensed retail, museum stores and premium brand programs often tighten to AQL 1.5 major and 2.5 minor. Appearance-critical commemorative sets may justify Level III or 100% sort on front-surface cosmetics only, with random sampling for dimensions and count. Whatever level you choose, the defect wording has to be objective. If the standard says color should look nice, the result will become an argument rather than a decision.

5) Control the bulk run with product-specific checks

Different products fail in different ways, so process checks should match the product. Enamel pins fail in fill level, over-polishing, color contamination, trapped dust in epoxy and post alignment. Keychains fail at jump-ring welds, split-ring spring temper, swivel function and plating wear on moving parts. Challenge coins fail at relief loss after polishing, uneven antique wash, sharp edge breaks and hairline scratches on mirror fields. Magnets fail at weak bond, warped metal plate, incorrect magnet placement and transit damage from metal-to-metal contact.

Use simple in-line checks that operators can actually maintain. Measure 5 pieces per SKU every 2 hours for size and thickness. Check hardware placement on the first 20 pieces of each shift and after any setup change. Run a 90-degree twist test on 3 pin-post samples per lot; reject if the post loosens, spins or separates from the solder pad or cast boss. For glued magnets, require at least 24 hours cure before final packing and perform manual peel testing on 3-5 pieces per lot. For keychains with blister or retail card packs, perform a 1 metre drop test on 1 packed unit per sampled carton set to expose weak blisters, denting or accessory movement.

Tolerances should reflect build method. Photo-etched brass can hold finer text and cleaner edges than standard stamping, but the visual effect is flatter. Die-cast zinc alloy handles openwork, sculpted forms and variable thickness well, but large flat back areas may show slight undulation or sink, especially above 60 mm. Functional pin bars on brooches and badges should normally stay within +/-1.0 mm of approved position; beyond that, the item may rotate or sag during wear. For challenge coins, rim width and edge reed definition should also be checked against sample because these details are often softened by aggressive polishing.

6) Inspect packed goods, not just loose pieces

A lot can pass cosmetic QC and still fail the buyer if the packing is wrong. Final inspection must cover both product quality and pack-out accuracy. Verify count per polybag or backing card, inner pack quantity, carton quantity, gross weight, net weight, carton dimensions, barcode readability and shipping mark placement. For standard manual warehouse handling, many importers set a carton gross-weight cap of 12-15 kg; above that, handling complaints and corner-crush rates usually rise.

For event distribution, retail fulfillment or multi-customer drops, assortment accuracy can matter more than one minor backside scratch. Mixed promo sets should be checked SKU by SKU to confirm correct combination, insert card version, orientation and label. Also inspect transit interaction: pin points should not mark adjacent cards, keychain rings should not dent acrylic or PVC inserts, and polished coins should be individually bagged or tissue-wrapped if packed next to harder metal components.

If you use a third-party inspector, send the defect catalogue before they arrive. Otherwise they may apply generic giftware standards and miss category-specific risks such as weak magnet bond, incorrect brooch orientation, poor split-ring temper or an antique finish that does not match the approved sample. Ask for inspection photos showing random carton selection, open-carton counts, measurement method, lux condition for cosmetic review and examples of both accepted and rejected defects.

• Verify AQL result by shipment lot size and by SKU, not just by total PO quantity
• Check piece count per pack, inner box count and full-carton count
• Confirm carton gross weight, dimensions and shipping marks against the booking plan
• Test hardware function: clutch grip, pin-post security, swivel rotation, split-ring closure and magnet bond
• Inspect front, back and sidewall finish separately because side coverage is often missed on plating
• Retain at least 3 approved bulk-run samples per SKU for reorder reference and claim support

7) Close the order with a reorder-safe archive

QC is not finished when the cartons ship. Reorders fail because the original decisions were never archived in a reusable format. Save the final approved artwork, mold layout, sample photos, measurement record, plating callout, packaging specification, carton specification, defect catalogue, in-process check notes, inspection report and any approved deviations. If slight backside swirl or minor antique variation was accepted on the first run, document it clearly so it does not become a dispute on the next PO.

The archive should also capture the commercial conditions that influenced quality and repeatability. Record actual sample lead time, actual bulk lead time, mold fee, rush surcharge, final overrun or underrun, scrap rate if disclosed, and any process issue that changed cost or schedule. Example: a high-polish antique silver coin with deep 3D relief usually needs slower polishing, more hand sorting and lower daily throughput than a flat 2D die-struck coin. Its FOB price at 300 pcs may not repeat at the same level under a compressed timeline or a different finish target.

Keep a short supplier scorecard after the first order. Score quote accuracy, process-fit advice, sample adherence, mass-production consistency, packing accuracy, corrective-action speed and communication. Then share that scorecard and the spec archive before the next RFQ. That turns the reorder into a controlled repeat instead of a new experiment with the same hidden handoff risks.

What to change on your next RFQ

Rewrite the next RFQ as a QC handoff document. Add dimensions, tolerances, Pantone references, line-width limits, plating notes, defect definitions, AQL, packaging counts, carton limits, MOQ tiers, USD FOB breaks and stage-by-stage lead times in working days. Ask each supplier to mark exceptions directly on your spec sheet rather than hiding them in a clean-looking quote.

At sample stage, approve only what you can measure or classify. Turn that sample into a one-page QC packet, require a first-article hold point after the first completed units, control the bulk run with product-specific checks, and inspect packed goods as carefully as loose pieces. That discipline usually costs less than one remake, one missed event date or one distributor return cycle.

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