I stood on bare concrete in a 900-employee greenfield plant and someone told me the labour rate made it all worthwhile. That was the easy sentence. The hard one came six months later, when we were still building calibration chains, writing control plans, training internal auditors, and chasing document approvals through a system that did not yet exist. The hourly wage was attractive. The fixed cost of standing up IATF 16949 from nothing was not.

Vietnam is being framed as the next EV manufacturing frontier. The headlines read like economic development copy. Strip the framing and the problem has nothing to do with industrial policy. It has everything to do with the fixed, immovable cost of building quality systems from a standing start.

The systems are the expensive part

When a government announces EV manufacturing ambitions, the press counts tax incentives, factory square footage, and projected headcount. Nobody counts the qualified internal auditors who do not yet exist in the labour pool. Nobody prices the calibration laboratories that need ISO 17025 accreditation before a single PPAP package can be signed off. Nobody budgets for the eighteen months it takes to build a PFMEA library that reflects the actual process rather than a template downloaded from a consultant's website.

The labour arbitrage is real. I will not deny that. When I built the QA/QC department for the SNOP greenfield — 900-plus employees, automotive Tier 1 — the wage structure was favourable. But I spent more time and money on quality architecture than on any single piece of production equipment. Calibration registers. Gauge R&R studies before the first serial run. Training programmes for inspectors who had never seen a control plan. A document control system that could survive a customer audit without anyone scrambling to backfill records on a Friday night.

Greenfield is greenfield

Here is the uncomfortable observation. Whether you build a plant in Vietnam, Poland, or South Carolina, the fixed cost of a compliant quality management system is roughly the same. IATF 16949 requires the same clause-by-clause implementation. VDA 6.3 demands the same process audit rigour. AS9100 — if you are running parallel aerospace supply chains — expects the same traceability and risk-based thinking. The cost of building these systems from zero is a function of the standard, not the geography. You can pay your operators less. You cannot pay less for the systems that govern how those operators work.

At SNOP, I stood up IATF 16949 and VDA 6.3 from that same bare concrete. Training, certification, time before people were productive — it exceeded equipment spend in the first year. We hit a 70% defect-cost reduction within cycle, but only because the quality architecture was designed and funded before volume. Not after. Not in parallel. Before.

Most organisations that fail at greenfield quality treat the management system as something that will emerge naturally from production scaling. It will not.

Distance amplifies defects

There is a comfortable assumption buried in offshoring decisions: that distance provides a buffer. It does not. Distance extends the detection-to-correction cycle, and every additional day between a defect's introduction and its discovery multiplies containment cost.

A field failure in California costs exactly the same whether the part shipped from Hanoi or Stuttgart. What changes is the timeline. A supplier thirty kilometres away can have a containment team on your dock in four hours. A supplier eight time zones away needs a week — if their 8D capability exists at all, and in greenfield operations, it frequently does not.

The further the factory, the longer the defect travels before anyone catches it — and the customer always receives it at full price.

I have run de-escalations where the root cause was geographic and organisational, not technical. The part was wrong because the process was wrong, and the process was wrong because the quality system was stood up under pressure, after volume, with untrained auditors signing off on control plans they did not write. Scrap, containment, expedited freight, customer line-down charges — it dwarfed whatever was saved on hourly wages.

Ford's CEO recently said the company is not satisfied with its quality wins and launched a campaign to win back customers. That is a domestic OEM with decades of quality infrastructure still struggling. Now imagine the same pressure on a greenfield operation three continents away, with a quality department that is six months old and still writing its first revision of critical procedures.

Where low-cost wins and where it collapses

Low-cost manufacturing wins when the quality architecture is designed, validated, and transferred before the first serial part runs. The SNOP greenfield succeeded — zero critical customer escalations within the first quarter, 98% customer satisfaction — because APQP was treated as the project, not as paperwork attached to it.

It collapses when leadership looks at the labour rate, runs the spreadsheet, and treats the quality system as a line item that will resolve itself through operational maturity. Operational maturity is a product of the quality system. Not the other way around.

The wave of AI-assisted quality tools being marketed to manufacturers is telling. The technology is genuinely useful — I build multi-agent AI systems myself. But the framing reveals the problem: organisations want tools to compensate for systems they never properly built. AI-driven defect detection layered on a weak APQP process is an expensive alarm bell on a house with no locks.

Key takeaways

  • The fixed cost of building IATF 16949, VDA 6.3 and APQP systems from greenfield is geography-independent — you save on labour, not on compliance architecture.
  • Defect cost does not decrease with distance. Longer supply chains extend detection-to-correction cycles, and field failure costs are identical regardless of origin.
  • Greenfield quality succeeds when the management system is funded and designed before volume — not alongside it, and never after.
  • Treat APQP as the project itself, not as documentation attached to the project. Organisations that invert this priority spend the difference on containment.

Vietnam's EV ambition will live or die on the same thing every greenfield automotive operation lives or dies on: whether someone is willing to pay the full, non-negotiable price of a compliant quality system before the first part ships. You can offshore the assembly line, the welding robots, the paint booth, the workforce. You cannot offshore the APQP. It travels with the part, it lives in the process, and the customer pays for its absence every single time.