"It was painful for us." Ford's chief executive said it about a quality disaster that shifted product plans, and the industry heard candour. I hear an autopsy. I have yet to meet the failure that surprised the plant. The sorting pen running on third shift, the air freight booked as routine, the credit quietly deducted from the next invoice – the shop floor could recite the symptoms months before anyone with a board badge felt a thing. Pain reached altitude only after it had finished eating the warranty reserve and moved the launch plan. By then it had been sitting in the ledgers for quarters. Nobody had added it up.

Pain is a lagging indicator

Executives feel quality last. Not from carelessness – the invisibility is structural. Warranty sits with finance, inside a reserve trued up quarterly. Containment and sorting hide in manufacturing overhead, one line among dozens. Expedited freight belongs to logistics, where it competes with weather and port strikes for blame, and customer credits get netted against revenue as a commercial matter. Four cost centres. Four owners. Four languages. Each is survivable on its own, which is the whole problem. A €180,000 sorting operation is a rounding error in overhead; the same figure repeated across twelve months and three product families is a strategy event. The sum that would make a board wince exists nowhere, because nobody owns it. Aerospace runs the same experiment with a longer fuse and larger invoices.

Defects are reported the day they are found. Their cost is reported the day it can no longer be absorbed.

Until someone aggregates the cost, quality travels upward as technical narrative – 8D reports, containment confirmations, closure evidence – while money travels as budget variance. Two reporting systems. The boardroom is fluent in only one of them. The press conference is what happens when the two finally collide in public.

Report failure like revenue

The fix is unglamorous. Cost of poor quality, monthly, by product family, with a named owner, printed in the same pack as revenue and walked past the people who control the budget. That is the entire mechanism. At SNOP, where I built the quality function of a 900-plus employee greenfield plant, the 70% cut in defect costs did not start with better tools – the tools were already on the floor. It started with one chart the CFO could not look away from, because for the first time the money had a shape. Once it did, QRQC stopped being a workshop ritual: every A3 closed a visible hole in a visible number, and 98% customer satisfaction followed the arithmetic. The visibility came before the improvement, not after.

Witte Automotive taught me the same mechanism at component scale. QRQC and A3 cut failure costs substantially, but only once we had aggregated sorting, expedited freight, customer line stoppage and credits into a single monthly total. A tool without a ledger behind it is theatre. The same tool with a ledger behind it moves headcount, capital and launch dates.

A confession buys ninety days

What follows a CEO-level quality admission is a campaign, and I have lived through enough of them to know the shape: war rooms, layered audits, gemba walks, a poster announcing that quality is everyone's job. A confession buys about ninety days of organisational attention before the next fire takes the oxygen. The test is not the intensity of the push. It is whether risk gets priced in euros before the launch decision, or scored in a PFMEA matrix after it.

I have sat in reviews where an engineer negotiated an occurrence rating from four down to three with the passion of a man haggling over carpets. The RPN dropped below threshold. Everyone went home satisfied. Not one person in the room converted a single number into money. The CFO has never once winced at an occurrence rating; he winces at €400 per unit of expected warranty. The gate question that protects plans is crude and binary: what does this failure mode cost if it escapes, what does preventing it cost, and which one are we choosing on purpose? Quality arriving as a score loses to schedule every time. Quality arriving as euros negotiates.

The customer, meanwhile, did the arithmetic long before the microphone was switched on. I once closed a quarter with zero critical customer escalations, and the mechanism was never perfection – it was that every known defect arrived at the customer with a cost, a containment date and a name attached. Customers forgive costed defects far faster than surprises. They keep the supplier who calls first. A confession delivered after the customer's own accountants have found the problem is not transparency; it is the invoice for lateness.

Key takeaways

  • Aggregate failure costs – warranty, containment, sorting, expedited freight, credits – into one monthly cost-of-poor-quality figure by product family, with a named owner. The number that never merges is the number that never escalates.
  • Put that chart in the same pack as revenue. QRQC, A3 and 8D become strategic only after the money has a shape; before that, they are workshop rituals.
  • Price risk in euros at the launch gate – cost of escape versus cost of prevention – instead of an RPN score after it. Budget holders negotiate with currency, not indices.
  • Call the customer before their accountants call you. A defect delivered with a cost and a containment date keeps the business; a surprise forfeits it.

Quality shifts corporate plans exactly twice: once when it breaks, and once when it is costed. The first shift is not ours to control. The second one is, entirely. Somewhere this quarter a plant manager will be told to support the quality push without ever being shown the sum, and that gap is the whole opportunity. Volunteer to be the number, not the narrative. Narratives about quality are always written by someone else – usually at a podium, usually too late.