Sourcing

Factory vs Trading Company for Custom Metal Promo Orders

When a low first quote turns into a costly program

A common 2026 sourcing failure starts with a strong first quote for 3,000 soft enamel pins or 1,000 challenge coins, then breaks down on the reorder. Rim height shifts from 1.5 mm to 1.3 mm. Polished nickel looks warmer on the second lot. Antique brass wipes darker. Clutch tension softens. Production moves from 12 working days to 19. The issue is rarely the spreadsheet price alone. It is whether the company quoting the order controls tooling, die striking, polishing, plating, color fill, assembly, packing, and the records needed to repeat the job.

That distinction matters most when the order is part of a program rather than a one-off. A 300-piece event pin can tolerate some variation. A quarterly brand program spanning 35 mm pins, 50 mm coins, 60 mm keychains, printed backing cards, and retail gift sets cannot. Once the requirement is the same Pantone match, same edge profile, same post placement, same barcode label, and same carton assortment on every run, the supply model starts driving real cost.

The practical question is not factory good, trader bad. The real question is which model gives tighter control over the specific failure points on your item: tooling ownership, metal thickness tolerance, plating consistency, assembly strength, pack-out accuracy, and accountability when the approved sample and the bulk lot do not match.

What changes between a factory and a trader

Decision point	Direct factory	Trading company	What buyers should ask
FOB unit price on standard metal items	Usually lower at 500+ pcs because there is no reseller margin. Example: 35 mm soft enamel iron pin, 1.5 mm thick, 2 colors, butterfly clutch: USD 0.29-0.44 FOB at 1,000 pcs. 50 mm zinc alloy keychain with split ring: USD 0.68-0.98 FOB.	Often similar on simple SKUs or USD 0.05-0.18/pc higher depending on the service layer and upstream source.	Request FOB pricing at 100, 300, 500, 1,000, and 3,000 pcs with mold fee listed separately.
MOQ structure	Typical MOQ is 100-300 pcs per design for pins and keychains, 100-200 pcs for standard coins, and MOQ often resets by size, finish, and packaging version.	Can be more flexible by combining factories, mixing low-volume SKUs, or absorbing small overruns across categories.	Ask whether MOQ applies per design, per plating color, per card design, or per combined PO.
Sampling lead time	Usually 5-7 working days for standard pins, 7-10 for coins, 8-12 for spinner keychains or kitted sets. Artwork revisions often add 2-3 days each cycle.	Quoted similarly, but actual timing depends on how quickly upstream sample capacity is booked and whether revisions go back through the same workshop.	Ask who makes the pre-production sample and whether each revision restarts the clock.
Mass production lead time	Typically 10-15 working days for standard pins and keychains after sample approval, 12-18 for coins, 15-25 for bundled retail sets.	May look similar on paper, but scheduling control is weaker if production is split across workshops or packing is outsourced.	Ask who controls production booking and whether the ship date is written into the PI.
Reorder consistency	Stronger if molds, Pantone references, plating notes, hardware specs, and golden samples stay in one production file.	Weaker if the trader changes workshops, cannot retrieve tooling data quickly, or sources the reorder from a lower-cost alternate factory.	Ask where tooling is stored, how long it is retained, and how prior specs are locked for repeat runs.
Engineering feedback	Usually better on minimum line width, recessed depth, relief height, post spacing, split ring gauge, weld quality, and edge treatment.	Can be slower or less precise if technical questions pass only through sales staff.	Ask who reviews artwork for manufacturability before tooling starts.
QC visibility	Easier to get in-line photos, plating checks, assembly checks, count checks, and final inspection records.	Varies widely. Good traders inspect upstream to a defined AQL. Weak traders inspect only sealed cartons or rely on supplier claims.	Ask for IQC, in-process checkpoints, and final inspection standard such as AQL 2.5 major / 4.0 minor.
Problem resolution	Faster when the seller also owns production and can sort, rework, or remake without cross-party disputes.	Slower when responsibility is split between trader and workshop or when root-cause ownership is unclear.	Ask who pays for remake, sorting labor, and expedited air freight if approved specs are missed.

Where factory-direct usually wins

Direct factory sourcing is strongest when the item is process-sensitive: hard enamel badges with fine metal lines, challenge coins with deep dual-side relief and edge text, spinner keychains with moving parts, brooches with multiple rear posts, magnets with defined pull performance, and gift sets where insert position and count accuracy must repeat batch to batch.

Take a 45 mm die-struck brass badge with hard enamel, polished nickel plating, 1.8 mm nominal thickness, and two rear posts. A capable factory should confirm manufacturable limits before tooling starts. Typical numbers are minimum raised metal line width around 0.25-0.30 mm, minimum recessed color cell around 0.35-0.40 mm, finished thickness tolerance of plus or minus 0.10-0.15 mm, finished size tolerance of plus or minus 0.20-0.30 mm, and rear post location tolerance of plus or minus 0.5 mm. On polished hard enamel, surface flatness and polishing consistency usually matter more than simply specifying extra thickness.

Plating is another area where technical detail matters. Decorative nickel, black nickel, gold-color, and antique finishes on promo metal goods are thin cosmetic coatings, commonly in the 0.03-0.08 micron range for the visible decorative layer, often over a copper flash depending on base metal and finish route. Buyers should not overread a micron number in isolation. Lot-to-lot consistency depends more on polishing preparation, bath maintenance, rack loading density, current control, and comparison to an approved sample or retained golden sample under the same lighting.

Most failures are predictable before the mold is cut. Serif text on a 38 mm coin can lose legibility if relief depth drops below about 0.20-0.25 mm. A 60 mm brooch can rotate on fabric if the posts are too close together or offset from the center of gravity. A keychain jump ring can open in use if the wire gauge is only 1.2 mm when 1.4-1.6 mm is more appropriate for the part weight. A factory engineer should catch those issues during artwork review, not after sample approval.

Reorders are the other major advantage. When one factory retains mold number, vector file, base metal, nominal thickness, finish code, Pantone callouts, attachment spec, bagging method, carton count, and approved sample photos in one system, the second and third orders are less likely to become reinterpretations of an old message thread. On recurring programs, that discipline is often worth more than a nominal USD 0.05-0.10 per piece price gap.

Where a trading company earns its margin

A trading company can be the better choice when the job is coordination-heavy rather than technically difficult. If the program includes 200 PVC keychains, 150 woven patches, 100 sublimated lanyards, 300 soft enamel pins, printed barcode labels, custom backer cards, and split shipping to two destinations, one competent coordinator may remove more friction than four direct factory relationships.

This matters for lean brand teams, agencies, and event buyers whose internal cost of supplier management is high. Good traders can consolidate cartons, align artwork approvals across categories, standardize barcode and carton marks, combine low-volume SKUs that do not meet direct-factory thresholds on their own, and issue consistent commercial invoice and packing list data.

The trade-off needs to be explicit. Once a trader sources from multiple workshops, variation risk rises unless every spec is locked in writing. Base metal may shift from brass to iron. Antique silver may come back darker on the next lot. Backing cards may be cut at plus or minus 2.0 mm instead of plus or minus 0.5 mm. Inner assortments may be wrong even when total carton count is right. That may be acceptable for giveaways. It is usually not acceptable for retail-adjacent merchandise, membership programs, museum shops, or tightly controlled corporate gifts.

The better traders operate like supply-chain managers, not quote pass-throughs. They should still be able to provide exact MOQ by SKU, sample lead time in working days, upstream process route, AQL plan, carton dimensions, gross weight, and written remake responsibility. If they cannot give those details, the coordination value is probably too thin to justify the margin.

The hidden costs that do not show on the quote

Most buyers compare mold fee and unit price first. The larger losses usually come from delayed launches, remake freight, inconsistent reorders, manual resorting after arrival, and internal time spent fixing preventable issues. A quote that is USD 0.06 higher per pin can still be cheaper if it avoids one 2,000-piece remake or one emergency air shipment costing USD 900-1,800.

Consider a 1,000-piece order of 50 mm die-struck challenge coins in brass with antique finish and dual-side 3D relief. Supplier A quotes USD 1.34-1.48 FOB with a 14 working day lead time after sample approval. Supplier B quotes USD 1.18-1.28 FOB with an 18-25 working day window. On paper, Supplier B saves roughly USD 120-200. That saving disappears quickly if edge lettering softens because the die was recut without reference to the previous lot, the antique wipe lands visibly lighter, or 5-8 percent of pieces fail inspection for burrs, underfilled recesses, plating voids, or mixed counts.

For promo metal goods, a practical baseline is final inspection to AQL 2.5 for major defects and AQL 4.0 for minor defects, with tighter limits such as AQL 1.5/2.5 for retail sets or licensed merchandise. Major defects typically include wrong item, wrong finish, unreadable text, missing attachment, sharp burrs, detached hardware, exposed base metal on the display face, wrong insert count, wrong barcode, or mixed SKU in a sealed retail unit. Minor defects usually include light scratches on the reverse, tiny enamel dust points, slight polish haze outside the hero area, or small print drift on the backing card.

Packaging errors deserve more attention than they usually get. On mixed promo orders, landed problems are often assortment defects rather than product defects: 48 units packed in a tray that should contain 50, the wrong SKU in an inner bag, left and right variants mixed in one carton, or carton labels that do not match the pack list. Those mistakes can erase any quote advantage because they create manual rework after receipt.

Questions that expose the real supply model

• Who cuts the mold, who stores it, and how long is tooling retained after the last order: 12 months, 24 months, 36 months, or permanent with active business?
• What is the exact base material by item: iron, brass, zinc alloy, stainless steel, or aluminum, and what is the nominal thickness tolerance, such as plus or minus 0.10 mm or 0.15 mm?
• What is the MOQ per design, per finish, and per packaging version? Request pricing at 100, 300, 500, 1,000, and 3,000 pcs where relevant.
• What are the sample and mass-production lead times in working days for this exact construction, not for a generic pin or coin?
• What hardware spec is used: butterfly clutch, rubber clutch, deluxe clutch, post length, split ring wire diameter, lobster clasp size, magnet grade, or adhesive type?
• What decorative plating range do you target, and how do you verify finish consistency lot to lot against the approved sample or golden sample?
• What QC checkpoints exist before packing: raw material check, post-stamping check, post-plating check, post-color-fill check, assembly check, count verification, and final AQL inspection?
• If bulk goods deviate beyond approved tolerance, who pays for remake, sorting labor, inspection rework, and expedited replacement freight?

These questions work because vague sellers struggle to answer them with numbers. A disciplined factory or capable trader can usually respond with specifics such as: sample in 7 working days, mass production in 12-15 working days after approval, iron base 1.5 mm plus or minus 0.10 mm, split ring wire diameter 1.4 mm, 50 pcs per inner bag, 500 pcs per export carton, carton size 38 x 28 x 22 cm, gross weight 12.5 kg, and final inspection to AQL 2.5/4.0. That is procurement data you can compare directly.

How the answer changes by product type

Not every promo item carries the same sourcing risk. Standard soft enamel pins, simple bottle-opener keychains, and basic embroidered patches are relatively forgiving if artwork, finish, and packing are documented. Hard enamel badges with mirror-polish faces, spinner keychains, layered coins, brooches for lightweight garments, and gift sets with EVA or foam inserts are less forgiving because tolerance, assembly, and visual repeatability matter more.

A standard 35 mm soft enamel iron pin at 1.2-1.5 mm thickness with two spot colors and one butterfly clutch is usually a stable factory or trader item at 500-1,000 pcs, with FOB pricing commonly around USD 0.29-0.44 and sample lead time around 5-7 working days. A 50 mm brass challenge coin with deep relief, sandblasted background, and edge text is different. At 3.0-3.5 mm thickness, FOB pricing is more often USD 1.30-2.20 at 500-1,000 pcs, with sample lead time around 7-10 working days and tighter dependence on die wear, polishing pressure, and antique wipe consistency.

A 60 mm brooch with a 2.0 mm body, decorative top element, and two tie-tack posts needs balanced post spacing and a stable center of gravity or it will sag in use. A fridge magnet made from 0.8 mm steel with bonded ferrite should specify magnet coverage area and pull expectation. If the requirement is to hold two to three A4 sheets on a painted steel surface, the drawing should define magnet size, placement, and bond method. A spinner keychain should define rotational smoothness, acceptable axial play, and hardware retention so the approved sample does not spin freely while the bulk lot binds or loosens.

The more the item depends on metal forming accuracy, finish control, or secondary assembly, the more factory-direct sourcing tends to outperform a general trader. The more the order depends on cross-category coordination, low-volume consolidation, and split packing, the more a strong trader can justify the margin.

A practical 2026 decision rule and RFQ format

Choose direct factory when at least three conditions are true: you expect repeat orders, the item has tight dimensional or finish requirements, engineering review matters before tooling, the order value is high enough that a remake is expensive, or the goods will be sold or closely inspected by brand stakeholders. In practice, that often means 1,000+ pcs per design, retail-grade merchandise, membership items, museum products, and any order where plating tone, edge quality, or attachment reliability must match previous batches.

Choose a trader when quantities are fragmented across categories, your team needs service-heavy coordination, and moderate process variation is commercially tolerable. Even then, run the project with the same documentation discipline you would require from a factory.

A usable RFQ package should include item size in mm, target material, nominal thickness, finish, attachment or hardware type, color references, packaging method, target AQL, shipping window, destination, and reorder expectation. Ask every supplier to reply in one fixed structure: MOQ tiers, sample lead time, mass-production lead time, FOB price by quantity, mold fee, carton count, carton dimensions, gross weight, and written remake responsibility. That forces comparable quotes quickly and exposes which suppliers actually control the process.

For mixed programs across pins, keychains, coins, patches, and lanyards, the best answer is often a hybrid model: direct control of process-sensitive metal production combined with coordinated sourcing of lower-risk accessories under one documented QC and packing standard. The supplier label matters less than whether they can prove control from artwork review through final carton sealing.

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