Cost of poor quality: the number your P&L hides.
Most plants know their scrap number. Far fewer know their cost of poor quality — the full bill for everything that was not right the first time. In my 14 years running plants, that gap between the two numbers was usually three to five times bigger than the finance team assumed.
What cost of poor quality means
Cost of poor quality (COPQ) is the total cost a business carries because products and processes fail to meet requirements. It is not just scrap. It is the rework labour, the re-inspection, the warranty claims, the expedited freight to recover a late shipment, the customer who quietly stops ordering. COPQ is the money you would get back if everything ran right the first time.
The reason it matters: scrap shows up on a report, but the larger share of COPQ is scattered across operations, logistics and sales where no one adds it up. Put a single number on it and quality stops being a cost centre and starts being the biggest improvement opportunity in the plant.
The PAF model: where the costs live
The standard framework, the Prevention–Appraisal–Failure (PAF) model, splits all quality-related spend into four buckets:
| Category | What it covers | Examples |
|---|---|---|
| Prevention | Stopping defects before they happen | Training, FMEA, process control, error-proofing |
| Appraisal | Finding defects through inspection | Incoming inspection, CMM, FAI, gauging, audits |
| Internal failure | Defects caught before the customer | Scrap, rework, re-test, downtime, sorting |
| External failure | Defects that reach the customer | Warranty, returns, recalls, lost goodwill |
The first two are the cost of control — money spent on purpose. The last two are the cost of failure — money lost by accident. COPQ is that failure half.
The COPQ formula
Total Cost of Quality = Prevention + Appraisal + Internal Failure + External Failure
Both are usually reported as a percentage of sales, so they can be tracked over time and compared across plants.
Reporting COPQ as a share of revenue is what makes it land in a management review. "We scrapped 4,000 parts" means little to a CFO. "Poor quality cost us 4.2 percent of sales this quarter" starts a budget conversation.
The hidden iceberg
The classic mistake is counting only the visible costs. Scrap, rework and warranty sit above the waterline because they hit a ledger. Below the surface sits the larger mass:
- Excess inventory held as a buffer against unreliable quality
- Expediting and premium freight to recover defect-driven delays
- Lost capacity on machines re-running rejected work
- Engineering time spent on corrective action instead of new products
- Lost sales and customer churn — the hardest to measure and often the largest
Quality thinkers have long argued the hidden costs run several times the visible ones. You will never cost them to the rupee, but a defensible estimate beats pretending they are zero.
Worked example: a ₹50 crore plant
Take a mid-size precision component plant with annual sales of ₹50 crore. Pulling the year's numbers into the PAF buckets:
| Category | Annual cost | % of sales |
|---|---|---|
| Prevention | ₹20 lakh | 0.4% |
| Appraisal | ₹55 lakh | 1.1% |
| Internal failure (scrap ₹1 cr + rework ₹40 L) | ₹1.40 cr | 2.8% |
| External failure (warranty + returns) | ₹60 lakh | 1.2% |
COPQ = ₹1.40 cr + ₹0.60 cr = ₹2.00 crore = 4.0% of sales.
Add the cost of control and the total cost of quality is ₹2.75 crore, or 5.5% of sales. Notice the imbalance: the plant spends only ₹20 lakh on prevention while losing ₹2 crore to failure — a 1:10 ratio. That is the signature of a plant that inspects quality in rather than building it in, and it is exactly where the easy money sits.
Industry benchmarks
- Typical manufacturer: total cost of quality of 15–20% of sales, dominated by failure costs.
- Average performer: roughly 8–12% of sales.
- World-class: below ~5% of sales, with failure costs the smallest slice because prevention is funded properly.
COPQ also tracks with process capability and sigma level — a process running at higher DPMO and sigma level generates fewer failures and so a lower COPQ. The two metrics tell the same story in different currencies. If you want the statistical side, our guides on Cp and Cpk and OEE show where failure cost is born on the line.
How to reduce COPQ
The lever is the prevention-to-failure ratio. Move spend left in the PAF model and total cost falls, because preventing a defect is cheaper than scrapping, reworking or shipping it. Practical moves:
- Error-proof the top three defect modes rather than inspecting for them.
- Tighten incoming quality so supplier defects never enter your process — an AQL sampling plan sets the rules.
- Raise first-pass yield with control plans and capability studies on the critical characteristics.
- Kill transcription errors in inspection and FAI paperwork — a surprising slice of failure cost is rejected submissions and re-inspection caused by numbers copied wrong between drawing and report.
On that fourth point: a lot of appraisal and internal-failure cost hides in manual documentation. Re-typing 200 dimensions from a drawing into an inspection sheet breeds errors that fail FAIs and trigger re-work. CadNexa's auto-ballooning tool reads each dimension and tolerance straight off the PDF, cutting the transcription step — and the rejected-submission cost that rides on it — out of your COPQ.
Frequently asked questions
What is the cost of poor quality?
The total cost incurred because products and processes are not right the first time — internal failure (scrap, rework) plus external failure (warranty, returns, recalls).
What is the formula for COPQ?
COPQ = Internal Failure + External Failure. Total Cost of Quality = Prevention + Appraisal + Internal Failure + External Failure, usually shown as a percentage of sales.
What is a good COPQ percentage?
Many plants run a total cost of quality of 15–20% of sales. World-class operations hold it below about 5%, with failure costs the smallest slice.
How do you reduce COPQ?
Move spend from failure to prevention — error-proofing, first-pass yield, incoming control, and removing transcription errors from inspection and FAI paperwork.