The Procurement Manager's Guide to Magnet Sourcing

Why magnets aren't like your other commodities

Four structural facts separate magnets from the fasteners-and-castings playbook, and every recommendation on this page traces back to one of them:

• The defining properties are invisible at receiving. Grade authenticity, coercivity at temperature, and coating integrity can't be caliper-checked. You are buying the supplier's process, which means qualification and documentation carry the load that incoming inspection carries elsewhere — the premise of the supplier qualification guide.
• Supply is geographically concentrated. The large majority of sintered NdFeB capacity — and nearly all rare earth refining — sits in China, which has used export licensing as policy leverage. Tariff stacks and export controls are live variables in your landed cost, per the tariffs & landed cost guide.
• The raw material is an index, not a price. Neodymium, praseodymium, and dysprosium move like commodities because they are commodities. A fixed-price annual agreement on a rare earth part is someone's speculation — yours or the supplier's. The price volatility guide covers the mechanics.
• The drawing doesn't fully define the part. Grades are property windows, magnetizing fixtures are house-built, and two conforming suppliers can ship measurably different parts. That makes resourcing slower and riskier than it looks — the central argument of the second-sourcing guide.

None of this makes magnets unmanageable. It makes them a category where the standard three-bids-and-a-buy motion quietly accumulates risk that surfaces later as a field failure or a supply gap.

Where the money goes

A custom sintered magnet's price decomposes into a few levers, and knowing which one dominates your parts tells you where negotiation is real and where it's theater:

Structuring the supply base

The practical choice is between three structures — direct factory, trading company, and U.S. stockist / value-add supplier — each with a distinct risk profile mapped in the qualification guide. The portfolio logic for a typical spend:

• High-volume, stable, engineered parts: direct factory relationships earn their overhead. Budget for qualification depth, site-change notification clauses, and owning the import problem yourself.
• Mixed-volume production parts: a U.S. stockist with named upstream factories gives you inventory buffer, landed pricing, and an English-speaking engineering interface — ask the qualification questions one level up the chain.
• Prototypes, spares, and legacy trickle: stock-size supply wherever possible. Custom tooling for a 500-piece-a-year part is the classic category leak.

Whatever the mix, the one clause that belongs in every agreement: identify the manufacturing site, and notify before it changes. A supplier who can move your part between factories without telling you has made your qualification file historical fiction.

The paper to require

Documentation is your substitute for the inspection you can't perform. The hierarchy, from the quality documentation guide, in the order you should demand it:

Acceptance criteria make the paper enforceable: every requirement needs property + number + method + condition, or it's an implied requirement — and implied requirements lose disputes. The tolerances guide has the vague-to-enforceable conversion table worth attaching to your next PO template.

Owning the risk, not just the price

Price variance gets reviewed monthly; supply risk gets reviewed after it happens. Flip that. The magnet-specific risk register has four standing entries:

• Policy risk: tariff modifications and export-control rounds arrive with weeks of notice, not quarters. Assign ownership for tracking them and pre-agree who absorbs a mid-contract duty change — in writing, per the landed cost guide.
• Index risk: decide your exposure policy — pass-through with an agreed index and formula beats fixed pricing that one party will eventually regret. Opportunistic pass-through (fast up, slow down) is a scorecard item.
• Concentration risk: which parts are single-sourced through a single site, and which of those would stop your line? The scored worksheet in the second-sourcing guide tells you where dual qualification pays and where inventory is the cheaper insurance.
• Silent-change risk: the lot that's suddenly different because the factory changed alloy recipe, plating line, or site. Site-change clauses, PPAP logic, and periodic re-verification of samples are the controls.

The quarterly routine

The category stays quiet on a lightweight scorecard reviewed quarterly — the same four axes as the qualification guide's ongoing scorecard:

Annually, run the six-step review from the cost reduction guide: it works the specification side (temperature classes, tolerances, custom-to-stock conversion) before the commercial side, which is where the durable savings are.

Inheriting the category: a 90-day plan

• Days 1–30 — see the spend. Part list with grade, temperature class, coating, annual volume, supplier, and manufacturing site (ask — the answer quality is itself a data point). Flag single-sourced line-stoppers.
• Days 31–60 — check the paper. Pull the last CoC and material cert for your top parts by risk. Missing lot traceability or a decade-old FAI tells you where to start. Verify certificates against registrar databases while you're at it.
• Days 61–90 — fix the two biggest leaks. Typically: one over-specified part burning money every order (send the drawing out for a cost-down review) and one single-sourced part with no continuity plan (price dual qualification vs. a stocking agreement, then pick one).

Related: RFQ preparation guide · supplier qualification · second-sourcing magnet parts