Here is an uncomfortable observation: when a defence minister says quality must be "embedded into every stage" of a two-year indigenous aircraft programme, he is describing the finish line of a decade-long maturity curve and calling it a starting instruction. India has announced indigenous aircraft manufacturing within two years, with an explicit call for stricter quality checks in defence production. I have built against a clock like that. Greenfield site, 900+ employees, QA/QC department from bare concrete upward. I know exactly what 24 months buys you — and what it cannot buy.

The speech leaves out the expensive part. Not the aircraft. The maturity.

What a decree can install — and what it cannot

Ministerial intent is real and it matters. A mandate gets you policy documents, an organisational chart, budget lines, certificates. I would take every one of those on day one. At the SNOP greenfield plant, standing up the QA/QC department for over 900 people, the auditable layer — the quality manual, IATF 16949-conformant document control, the process maps — was the fast part. That layer arrives by signature.

The working layer does not. Control plans that operators actually follow. Reaction plans that trigger when a process drifts, before a defect ships. People who stop the line without first calculating whether their supervisor will punish them for it. That layer arrives only by repetition — thousands of cycles of escalation, review and closure, with leadership behaving the same way in week 80 as in week 8.

A certificate can arrive before the first real escape. It just can't answer for it.

AS9100 certification in 18 months is achievable — Vertical Aerospace just demonstrated that the certificate layer is purchasable with enough consultants. What it cannot purchase is what an auditor never sees: whether your reaction plan has ever been executed under pressure, at two in the morning, with a shipment waiting on the outcome.

The honest 24-month sequence

Here is the timeline I actually ran, translated to an aerospace start:

  • Months 1–6: Hire the bench and stand up document control. Not quality engineers first — inspectors, metrology people, a document controller who is genuinely feared. Greenfield sites cheat here by hiring planners before quality, because planners move metal. The shortcut costs roughly €40,000–80,000 per escaped defect when it surfaces at the customer, against maybe €200 caught in-house.
  • Months 7–12: Run your first PPAPs or first-article inspections and take your first escapes. You will take them. The point is taking them cheaply — internally, in QRQC within 24 hours, A3 for the systemic ones. Incumbents cheat differently: they assume experience transfers and skip PFMEA depth. It doesn't, and they shouldn't.
  • Months 13–24: Build audit cadence — layered process audits, VDA 6.3-style process discipline — and supplier depth. This is where defect cost actually collapses. We cut defect costs by 70% at that greenfield plant and reached 98% customer satisfaction, and I can tell you precisely when it happened: months 14 through 22, after the audit cadence and supplier development bit. Not before.

Two years, respected as a sequence, delivers structure, certificates and a department that knows its own processes. Two years spent skipping the sequence delivers the same certificates on top of unmatured processes — the most expensive combination there is, because now the paperwork testifies that nothing is wrong.

The supplier tier nobody schedules

National aircraft programmes do not lean on their own factory. They lean on a local supply chain that, in large part, has never made flight hardware. Traceability, special processes, first-article inspection, NDE — new disciplines for a tier that has spent its life serving automotive or general industry. I have sat on the supplier-development side through FOREAST with companies like ArcelorMittal, Meopta and AVX Kyocera, and later led quality across a 2,000+ multi-site workforce. Supplier development — capability walks, joint A3s, VDA 6.3 audits that develop rather than merely gate — runs on a longer clock than the aircraft programme sitting on top of it.

Watch the news flow and you see the problem live. Hyderabad is pushing an NDE skills drive because aerospace jobs are expanding. Inspectors being trained in non-destructive evaluation — penetrant testing, ultrasonics — are exactly the bench a flight-hardware supply chain needs. But that recruitment is happening after the two-year deadline was announced, not before. The deadline was set before the capability existed. That is backwards, and it is the single most reliable predictor I know of a programme that will end up buying maturity retroactively, at retail price.

Key takeaways

  • Decrees install the auditable layer — policy, org charts, certificates. The working layer — control plans, reaction plans, fearless escalation — only arrives through repetition and cannot be scheduled.
  • A realistic 24-month build: hire inspectors and document control first, run early PPAPs and absorb cheap internal escapes by month 12, then spend the second year on audit cadence and supplier depth, where defect cost actually collapses.
  • Supplier development runs on a longer clock than the aircraft. Training the NDE bench after setting the deadline means paying for maturity later, at escalation prices.
  • Certificates earned before the first real escape certify paperwork, not capability — the escape will audit the certificate, not the other way round.

What I would tell that programme office

Spend the two years building the system and the aircraft follows. Spend them building the aircraft and the system bills you later — in customer escalations, in escapes at the customer's dock, in certificates wallpapering over processes nobody has stress-tested. I have watched both paths run, in automotive plants and in aerospace supply chains, and the invoice for skipped maturity always arrives. Always with interest. The minister is right that quality belongs in every stage. He should also be told what every plant manager knows: embedding it takes longer than announcing it, and the two-year clock starts measuring maturity only after someone has built the department that can hold it.