Design to Cost · Toolkit
Cost of Poor Quality
Add up scrap, rework, warranty, returns and firefighting — the hidden factory is usually the cheapest one to close.
- Time45 min
- FormatSolo
- StageToolkit
Cost of Poor Quality: what it is and why it works
Cost of Poor Quality (COPQ) adds up what the organization spends because work is not done right the first time: internal failures such as scrap and rework, external failures such as warranty, returns, field repairs and complaint handling, and the firefighting around them, including expedited freight, overtime and management time. In the prevention-appraisal-failure view associated with Feigenbaum, these failure costs sit alongside appraisal costs (inspection and testing) and prevention costs (training, design reviews, process capability work). The method collects a year of data per category, expresses the total as a share of revenue, breaks the largest category down by cause and funds improvement projects against the biggest causes first.
COPQ works because it translates quality problems into the language management uses to allocate money. Scrap and rework are often partly absorbed in standard costs and overtime, warranty sits in a separate account, and firefighting is spread across salaries, so no single report shows the total. Once summed, the hidden factory frequently turns out to be a large and cheap source of savings compared with redesign or supplier negotiation. Crosby's argument that quality is free rests on the same logic. COPQ links to Variance Analysis, which explains cost overruns, to Activity-Based Costing, which captures firefighting time, and to Design-for-Cost Rules, which prevent recurring failure modes from being designed in again.
What you need
- One year of scrap, rework, warranty, returns and complaint data
- Labor and overtime records linked to rework and firefighting where possible
- Expedited freight and customer penalty or chargeback records
- Revenue for the same period
- Defect and failure cause data from quality systems
What you get
- COPQ by category and in total, as a share of revenue
- Pareto of causes within the largest category
- A ranked list of improvement projects with expected savings
- A baseline for tracking COPQ over time
When to use it
When quality is seen as a cost center instead of a cost source.
How to do it, step by step
- Define the categories: internal scrap, rework, warranty, returns, complaint handling.
- Collect one year of data per category, in euros.
- Compute total COPQ as share of revenue.
- Break down the largest category by cause.
- Fund improvement projects against the biggest causes first.
Worked example: Sizing the hidden factory at a precision machining supplier
Illustrative scenario — figures are realistic but not from a real company.
A machining supplier to aerospace and defense customers, with revenue around $38 million, believed its quality costs were under control because the scrap account showed about $410,000 a year. The plant manager asked for a full COPQ estimate.
- The quality engineer gathered one year of data: scrap $410,000, rework labor $290,000, customer returns and replacement parts $180,000, complaint handling and corrective action reports $95,000, and expedited freight to recover late shipments $120,000.
- The total, about $1.1 million, was close to 3% of revenue, more than two and a half times what the scrap account alone suggested.
- Breaking down scrap by cause, one titanium bracket family accounted for close to half, mostly first-piece losses after setups on a five-axis cell.
- A project was funded to introduce presetting of tools and fixture offsets and a first-article probing routine on that cell.
Result. Titanium scrap on the bracket family fell by about 60% within two quarters, saving roughly $120,000 a year, with knock-on reductions in expedite freight. COPQ became a monthly metric in the plant review, and the next project targeted rework on anodized parts.
Common pitfalls and how to avoid them
- Counting only scrap, which misses rework, warranty and firefighting.Define categories covering internal failure, external failure and firefighting, and estimate each even if data is imperfect.
- Waiting for perfect data before reporting anything.Start with reasonable estimates, label them, and improve data collection as projects progress.
- Reporting the total without breaking it down by cause.Pareto the largest category by product, process and cause to find fundable projects.
- Cutting appraisal costs, such as inspection, to reduce COPQ quickly.Reduce failures first through prevention; removing inspection without fixing causes usually shifts cost to external failures and can create safety or compliance risk.
Frequently asked questions
What are the four categories of quality cost?
The classic prevention-appraisal-failure model uses prevention costs (training, design reviews, process planning), appraisal costs (inspection, testing, audits), internal failure costs (scrap, rework, downtime from defects) and external failure costs (warranty, returns, complaints, field repairs). Cost of poor quality usually focuses on the failure categories, sometimes including excess appraisal.
How do you calculate the cost of poor quality?
Collect costs for each failure category over a defined period, typically a year: scrap, rework labor and materials, warranty claims, returns, complaint handling, expedited shipping and related overtime. Sum them and divide by revenue for the same period. Document estimation methods for categories without direct accounts so the figure can be repeated consistently.
What is the hidden factory?
The hidden factory is the set of activities that exist only to deal with problems: reworking parts, sorting, re-inspecting, expediting, handling complaints and rescheduling. It consumes capacity, people and money but does not appear as a separate line in the accounts, which is why cost of poor quality analysis is needed to reveal it.
Origin
Cost of poor quality — Armand V. Feigenbaum, "Total Quality Control", 1951; Crosby, "Quality Is Free", 1979.
Related methods
- Cost Variance AnalysisExplain every gap between estimate and actual: which assumption was wrong, and what does that teach the model?
- Activity-Based CostingAllocate overhead by the activities that actually consume it, not by a flat percentage of direct cost.
- Design-for-Cost RulesWrite ten costing rules your designers can check in seconds: standard tolerances, preferred materials, no…
More in “Toolkit”
Cross-cutting methods: curves, complexity, quality and estimating tools.