TL;DR:
The criteria that actually matter when picking a precision metal stamping supplier are an industry-matched quality cert, explicit tooling ownership language, and real engineering input at the DFM stage. Evaluating on quoted piece price and press count tends to blow up around month 18, usually in the form of tooling disputes or quality drift. Switching costs on a transferred die regularly run past $50,000 once you count the downtime.
1. Why Most Stamping Supplier Evaluations Fail
Most procurement teams score stampers on two numbers: the PPAP quote and the piece price. It works fine for about a year and a half. Then you hit a capacity ceiling, or Cpk starts drifting, or someone realizes the CAD files aren’t actually yours. The global metal stamping market is north of $218 billion, and the spread in supplier maturity inside that number is enormous.
Hitting ±0.005″ on a first article tells you almost nothing about whether a shop can hold the process for 400,000 parts. And when you decide to move a tool because the shop can’t, you’re looking at 12 to 16 weeks of lost production and a stack of engineering charges nobody warned you about. The ten criteria below are what I actually use on a scorecard.
2. Quality System Certifications — Match the Cert to Your Industry
Match the cert to the industry: IATF 16949 for automotive, AS9100 for aerospace, ISO 13485 for medical. ISO 9001:2015 is the floor for general industrial, but it doesn’t carry the risk management or traceability teeth the regulated sectors need.
Certs also move. IATF published the Rules 6th Edition and it became mandatory for all audits starting January 1, 2025. Aerospace stampers are mid-transition from AS9100 to IA9100 right now, which is worth asking about directly because some shops are further along than others. Medical stampers need ISO 13485:2016 plus FDA 21 CFR 820.50 purchasing controls.
| Industry | Required Cert | Audit Frequency | Common Add-ons |
|---|---|---|---|
| Automotive (Tier 1/2) | IATF 16949 | Annual surveillance | VDA 6.3, CQI-9 (Heat Treat) |
| Aerospace & Defense | AS9100 (IA9100) | Annual surveillance | ITAR Registration, Nadcap |
| Medical Device | ISO 13485 | Annual surveillance | FDA Registration |
| General Industrial | ISO 9001 | Annual surveillance | ISO 14001 (Environmental) |
3. Process Capability — Looking Past Tolerance Claims
Any shop can hold tolerance on a 30-piece sample. The question is whether they can hold it in production, which means SPC data, not first-article reports. Ask for actual Cpk and Ppk from running jobs.
AIAG PPAP 4th Edition sets the number everyone references: Cpk > 1.67 for preliminary (short-term) studies, Cpk > 1.33 for ongoing production. A 1.33 puts the process mean at least four standard deviations off the spec limits. Fine in theory. The real tell is whether the shop can produce the chart without a three-day scramble.
Ask the Quality Manager:
- “How long do you retain SPC data for critical characteristics?”
- Do you run in-die sensing for slug pulling and misfeeds?
- Can I see a sanitized Cpk report for a part with similar thickness and tolerance band?
4. Tooling Ownership, Transfer, and the Exit Clause
This is the one that bites people, so I’m going to spend some time here.
Tooling ownership means who legally owns the die after it’s built. Normally the buyer pays, buyer owns. In practice, contract language and the physical reality of the tool sitting in someone else’s building hand effective control to the shop. Browse the manufacturing subs on Reddit for ten minutes and you’ll find the same dispute replayed over and over: buyer wants to move the tool, shop sits on the CAD, extraction fee suddenly appears, lead time to move stretches past the point where it’s cheaper to just build a new die. Which, of course, is the point.
A real supplier evaluation includes a review of the tooling bailment agreement. That’s the legal document establishing the shop holds the tool in trust and does not own it. On top of that, file a UCC-1 financing statement in the supplier’s state. Public declaration of ownership, and it matters if the shop goes under.
A few contract items to actually negotiate, not just read:
- CAD data rights. The contract should explicitly say you own the 3D tool design files. Not just the steel. The files.
- Tool life guarantees with a number. “Industry standard tool life” is not a number. “1,000,000 strokes before the supplier is responsible for major rebuild” is a number.
- Transfer prep fees, defined up front. Anecdotally I’ve seen prep fees land somewhere in the low-single-digit thousands per die for a straightforward progressive, more for anything with a lot of stations or sensors. Get the number in writing before you sign, not after you’ve decided to leave.
- Right of access. If you need to send your own toolmaker in to inspect the die, that needs to be a contractual right.
I still push buyers to overspecify this section even when the shop pushes back, because the shops that push back hardest are usually the ones you’d most want to be able to leave.
5. Engineering Depth — Can They DFM, or Only Quote?
Quote-the-print shops take your file, assume it’s correct, and price the geometry. An engineering partner runs DFM on it and comes back with a shorter list of stations, a better strip layout, or a note that your corner radius is going to crack on the second hit.
Look at the software stack and the in-house tool design bench. The serious shops run AutoForm (R11 dropped mid-2024 with reworked springback compensation) or FTI FormingSuite to simulate flow, splits, wrinkles, thinning. Early DFM involvement commonly knocks 5-15% off material consumption, which is where your actual piece-price savings live over a program.
6. Equipment, Press Mix, and Hidden Capacity Realities
“We have 50 presses” is not information. You need tonnage distribution, press age, and the technology mix. A floor of 30-year-old mechanicals is going to struggle with modern AHSS geometries in ways that are not going to show up in the quote. Shops running direct-drive servo lines from Aida, Minster, or Komatsu can program slide motion, slow through the draw, speed through the return, and pick up 30-50% in stroke rate.
Ask for OEE. A shop at 85% is running real processes. A shop at 40% is going to miss ship dates and blame the material.
| Press Type | Typical Application | What to Ask About |
|---|---|---|
| Mechanical (Flywheel) | Blanking, simple progressive dies | Clutch/brake maintenance records |
| Servo | Deep draw, high-strength steel, complex forming | Stroke profile programming capability |
| Transfer | Large parts requiring multiple independent stations | Automation/transfer mechanism speed |
Obviously you’re not spec’ing a servo program for a plain blanking job, but you’d be surprised how often someone tries.
7. Supply Chain Transparency Down to the Coil
Transparency here means knowing where the coil comes from. A shop living off spot-market import steel is carrying more risk than one with mill-direct allocation, and that risk shows up as missed deliveries the first time tariffs or capacity move. US raw steel capacity utilization was running around 76-77% in late 2024 and early 2025, so this isn’t theoretical.
Require MTRs with every shipment. And look at the secondary ops network. Zinc plating, heat treat, e-coat — in-house or outsourced? If outsourced, you need to audit the subtier too, because their quality is now your quality.
8. Financial Health Signals You Can Check Without an NDA
A broke shop defers maintenance on your tooling and slow-pays its steel supplier. You can get a surprisingly clear read before you ever sign an NDA.
90-minute financial due diligence:
- Pull a D&B report. PAYDEX of 80 means paying exactly on terms. Below 70 means a trend of 15+ days late. Cash flow is tightening.
- Run a UCC lien search through the Secretary of State in the shop’s state. A cluster of recent liens on press equipment is a flag.
- Check LinkedIn headcount over 24 months. A 20% drop in a year usually precedes the kind of operational collapse that ends with your tool in a receiver’s warehouse.
- Look at the equipment nameplates when you tour. If nothing on the floor was bought in the last decade, that’s a capex story and it’s not a good one.
9. Facility Audit — What to Actually Look At Beyond the Tour Route
Supplier tours are sanitized. You see the cleanest aisle and the newest press. The information is in the places they don’t walk you through. VDA 6.3 (2023 edition), which the automotive guys lean on heavily, puts a lot of weight on non-conforming parts handling and material flow for exactly this reason.
Where to actually look:
- Scrap bins. Red bins overflowing? Mixed with good parts? That’s a defect-escape risk, full stop.
- The toolroom. Organized bench, labeled components, or a pile? A messy toolroom means delayed die maintenance, which means your die.
- Operator certification boards. Pull up to a workstation and read the training matrix. If the operator’s cert expired in 2023, the board is decoration.
- Safety stock. Physically count what they claim to be holding for your program. Don’t take the number from the ERP screen.
10. Communication Cadence and Program Management
The last one decides how painful launch is going to be. You need a dedicated Program Manager running the APQP phases. Without one, engineering questions bounce between your team and the toolroom and the answers show up a week late and partially wrong.
Also look at the IT side. High-volume production needs EDI. In automotive and industrial that means ANSI X12, specifically the 830 (Planning Schedule) and 862 (Shipping Schedule) transaction sets. Shops still doing manual email releases will stock you out. It’s not a question of if.
Scorecard weighting, rough starting point:
- Automotive: quality certs and Cpk at 40% of total.
- Medical: supply chain traceability and facility audit at 40%.
- Aerospace: engineering depth and material certs at 40%.
Adjust from there based on how much your line-down hour actually costs you.
FAQs
1. What’s the difference between a precision stamper and a general metal fabricator?
A precision stamper runs hard tooling (progressive or transfer dies) in mechanical or servo presses, high volume, tight tolerances, often ±0.001″. A fabricator runs soft tooling — lasers, brakes, turret punches — for low-volume, high-mix work. Different business model, different cost structure.
2. Who owns the stamping tooling — the buyer or the supplier?
The buyer owns the tooling once it’s paid for, assuming the contract has a clean bailment agreement. The supplier physically holds it, which is why a UCC-1 filing matters — it protects your ownership against the supplier’s creditors if things go sideways. Without the UCC-1, your tool is just another asset in the bankruptcy estate, and you’ll be standing in line behind the secured lenders trying to argue that the die with your part number on it is actually yours. I’ve watched that argument fail more than once.
3. How long does it take to qualify a new precision metal stamping supplier?
Three to six months from RFI through PPAP, assuming nothing weird happens.
4. Should I single-source or dual-source a stamped component?
Single-sourcing is cheaper upfront — one die, one PPAP. Dual-sourcing cuts supply risk but doubles your tooling capital and your qualification work. Only dual-source if a line-down event costs more than the second tool.
