3.6 Cost of Poor Quality and Hard versus Soft Savings
Key Takeaways
- The four COPQ categories are prevention, appraisal, internal failure, and external failure; prevention and appraisal are costs of conformance, the two failure categories are costs of non-conformance.
- External failure cost is the most expensive category because it includes warranty, recall, liability, and lost customers.
- Hard savings appear as a measurable change in the profit and loss statement; soft savings are cost avoidance or capacity release that never reaches a financial statement line.
- The 1-10-100 rule expresses the escalating cost of a defect found at prevention, at inspection, and at the customer.
- COPQ commonly runs 15% to 25% of revenue in an organization operating around three sigma, and most of it is hidden below the visible scrap and rework line.
Defining cost of poor quality
Cost of poor quality (COPQ) is the cost that would vanish if every process performed right the first time, every time. It is a superset of scrap and rework: it includes the inspection you perform because you do not trust the process, the expediting you pay because the first shipment was wrong, and the customers who did not come back.
COPQ is the Black Belt's principal argument. It converts a quality problem into a financial problem and reframes improvement as recovering money already being spent rather than requesting new investment.
The prevention-appraisal-failure model
| Category | Definition | Examples |
|---|---|---|
| Prevention | Cost of preventing defects from occurring | Design reviews, DFMEA and PFMEA, supplier development, training, error-proofing, process capability studies |
| Appraisal | Cost of assessing whether output conforms | Incoming inspection, in-process inspection, final test, gauge calibration, audits, MSA studies |
| Internal failure | Cost of defects found before the customer receives the product | Scrap, rework, re-inspection, downgrading, downtime caused by defects, yield loss |
| External failure | Cost of defects found after delivery | Warranty claims, returns, field service, recalls, liability, concessions, lost customers |
Prevention and appraisal together are costs of conformance -- money spent deliberately to achieve quality. Internal and external failure together are costs of non-conformance -- money lost because quality was not achieved.
The classic relationship is that spending on prevention reduces failure cost by more than it costs, and that appraisal spending has diminishing returns because inspection cannot create quality it can only detect its absence. A plant whose quality budget is 80% appraisal and 5% prevention is paying to sort rather than paying to fix.
The 1-10-100 escalation
The cost of a defect grows by roughly an order of magnitude at each stage it escapes.
| Where the defect is caught | Relative cost | Why |
|---|---|---|
| Prevented at design or setup | 1 | Change a drawing, adjust a parameter |
| Detected internally at inspection | 10 | Material and labour already consumed; rework or scrap |
| Detected by the customer | 100 | Warranty, freight, field service, investigation, and reputation |
The rule is a heuristic, not a measured constant, but the ordering is reliable and it is why external failure dominates most real COPQ studies.
The iceberg: visible versus hidden COPQ
Most organizations measure only the top of the iceberg because those costs already have accounts.
Visible (measured today): scrap, rework labour, warranty claims, returns, inspection headcount.
Hidden (rarely measured): excess inventory held as protection against defects, expediting and premium freight, engineering time spent on corrective action, extra capacity built to absorb yield loss, order cancellations, customer defections, management time in quality escalations, and opportunity cost of capacity consumed producing defects.
Hidden costs typically exceed visible costs by a wide margin. In an organization running near three sigma, total COPQ commonly falls in the 15% to 25% of revenue range, while the visible portion booked to quality accounts might be 3% to 5%.
Hard versus soft savings
This distinction is named explicitly in the Body of Knowledge and is a reliable exam topic.
| Hard savings | Soft savings | |
|---|---|---|
| Definition | A measurable change in the profit and loss statement | Cost avoidance, cost reduction, or capacity release that does not change a P&L line |
| Test | Would the controller book it? | Real benefit, but no line item moves |
| Examples | Material cost eliminated, headcount removed or redeployed against an open requisition, scrap expense reduced, freight premium eliminated, warranty accrual reduced | Hours freed but not removed, capacity released without additional sales, avoided future costs, avoided potential fines, cycle-time reduction with no revenue effect |
| Reporting rule | Counts toward validated project benefit | Reported separately, never mixed into the hard number |
Two frequent errors:
- Counting freed labour hours as hard savings. Saving 0.4 of a person's time is only a hard saving if 0.4 of a person's cost actually leaves the organization -- through a removed position, an unfilled requisition, or eliminated overtime. Otherwise it is soft.
- Counting released capacity as revenue. Freeing 300 machine hours only becomes a hard saving if those hours are sold or if they let you cancel outsourced work.
Cost avoidance deserves its own note. Avoiding a future cost that has not yet been incurred is genuine value -- avoiding a recall, avoiding a capital expansion, avoiding a regulatory penalty -- but because the counterfactual cannot be observed, it is soft by convention. Report it, quantify it, label it.
Worked example
A moulding line produces 1.2 million parts per year. Baseline scrap is 4.2% at a material and conversion cost of $2.85 per part. Rework runs 1.1% at $1.40 per part. Premium freight to cover late shipments caused by yield loss runs $46,000 per year. Two inspectors at $58,000 fully loaded exist solely because the process is not trusted.
- Scrap: $1{,}200{,}000 \times 0.042 \times $2.85 = $143{,}640$ (internal failure)
- Rework: $1{,}200{,}000 \times 0.011 \times $1.40 = $18{,}480$ (internal failure)
- Premium freight: $46{,}000$ (internal failure, hidden)
- Inspection labour: $116{,}000$ (appraisal)
- Total identified COPQ: $324,120
A project that halves scrap and rework and eliminates premium freight yields $0.5(143{,}640 + 18{,}480) + 46{,}000 = $127{,}060$ of hard savings, because material and freight spend genuinely falls. Removing one inspector position saves a further $58,000 hard only if the position is actually eliminated; if the inspector is redeployed to another area that had an open requisition, it is still hard, but if they simply have less to do, it is soft.
A gauge calibration program and an incoming inspection department fall into which cost of quality categories?
A project frees 0.6 full-time equivalents of operator time. The operators remain employed in the same department with no reduction in overtime or headcount. How should the benefit be reported?
A plant spends 78% of its quality budget on inspection and test and 4% on prevention activities, and its external failure costs are rising. What does the prevention-appraisal-failure model predict?