Poor quality commonly consumes 15% to 20% of sales revenue at many manufacturers, while broader industry estimates place the burden at 5% to 30% of gross sales [S1][S3][S7][S8]. These are planning ranges rather than a substitute for a plant calculation.
For many manufacturers, quality-related costs run between 15% and 20% of sales revenue, so a plant with $80 million in annual shipments could face a potential poor-quality burden of $12 million to $16 million, but the defensible figure must be built from traceable costs such as scrap, rework, quality-stop downtime, warranty claims and returns [S1] [S2].
What Counts as Cost of Poor Quality
Cost of poor quality, or COPQ, is the money lost because work was not completed correctly. It includes internal failure costs, such as scrap, rework, re-inspection and quality-stop downtime, plus external failure costs, such as warranty claims, returns, recalls, expedited freight and lost orders [S2][S3].
The wider cost-of-quality model divides expenditure into four categories. Prevention covers training, maintenance and process design; appraisal covers inspection, testing, calibration and audits. Internal failure covers defects found before shipment, and external failure covers costs that reach the customer. Prevention and appraisal form the cost of good quality, while the two failure categories form COPQ [S1][S3].
This distinction matters for budgeting. A plant cannot label all inspection spending as waste, because appraisal is part of a complete quality system. It also cannot treat prevention as a free activity, because training, supplier qualification and equipment maintenance consume real resources [S2].
How to Build a Plant-Level Annual Figure
A defensible annual COPQ calculation is built line by line from the plant's systems of record rather than from a percentage benchmark.
The calculation should use fully loaded unit costs. Material, labor, machine time and energy already invested must be included, together with the cost of producing a replacement or repeating a failed operation. Internal failure records can be derived from scrap and rework counts or from first-pass-yield and rolled-throughput-yield data, while external failure records come from warranty, return and complaint systems [S3].
APQC benchmarks the measure as total annual cost of poor quality per $1,000 of revenue, which creates a comparable rate rather than a company-specific dollar estimate. A useful management dashboard can track that rate by line, shift, product family and failure type [S4].
Benchmark Range and Plant Reality
The 15% to 20% range is the most commonly cited benchmark for quality-related costs at many manufacturers. A broader estimate covering manufacturing and service operations puts COPQ at 5% to 30% of gross sales, with many manufacturers falling in the 10% to 20% band [S1][S7][S8]. Product complexity also changes the exposure: one 2026 manufacturing estimate gives a range of 5% to 25% of sales [S6].
These percentages are planning ranges, not a substitute for a plant calculation. For example, at $10 million of annual revenue, 10% equals $1 million, 15% equals $1.5 million, and 20% equals $2 million. At $100 million, the same percentages become $10 million, $15 million and $20 million. A plant with $80 million in shipments reaches 5.3% in the worked example because its measured failure cost is only $4.24 million [S2].
The wider spread matters because product mix, process capability, inspection strategy and customer obligations change the cost of failure. A plant making high-complexity assemblies with expensive materials and tight customer specifications carries a different exposure from one producing simple, low-value components. This is why the same benchmark should not be assigned automatically to every line [S6].
Internal and External Failures Compared
Internal failure is normally visible to production and maintenance, while external failure is often split across finance, logistics, customer service and sales. A practical comparison is: internal failure consumes material, labor, machine hours and capacity before shipment; external failure adds claims, returns, replacement freight, penalties and lost business after shipment; and both can remain hidden because no single cost center owns the complete problem [S1][S2].
Early detection can be financially attractive because it can prevent downstream failure costs, but the appropriate control point depends on the process, cost structure and risk of escapes. It is not a universal accounting rule to assume that earlier detection always costs less.
What Drives the Price of Quality Improvement
The purchase price of an inspection system is only one part of the decision. Material cost drives scrap and replacement exposure; labor cost drives rework, re-inspection and complaint handling; machine-hour and energy cost drive downtime; and compliance or contractual exposure raises the consequence of defects escaping the plant [S2][S3].
Volume affects the economics. A camera-based inspection system that replaces repetitive manual inspection can be justified when avoided scrap, rework and downtime exceed installation and operating cost, but the plant still has to price image processing, line integration, false-rejection risk and maintenance. The research provides no universal equipment price, so the defensible approach is to compare quoted capital cost with avoided annual failure cost from the plant's own records [S2].
Total cost of ownership must include installation, integration, operator training, calibration, consumables where applicable, software maintenance and planned downtime. Maintenance programs and error-proofing can prevent defects before they create rework or warranty exposure, while upstream process improvement may be more effective than adding another final inspection gate [S3][S5].
One Process Lab Pro calculation returns $756,000 in annual quality cost, including $558,000 in failure costs and a 74% failure share. Its category mix is 4% prevention, 22% appraisal, 50% internal failure and 24% external failure; within that example, $9.3 million in additional sales would be needed to recover the failure cost [S5].
Who Should Use the Benchmark and Who Should Not
The benchmark is appropriate for plant managers, quality leaders and finance teams establishing an initial COPQ baseline when transaction-level data is incomplete. It supports directional questions such as whether 15% of revenue is being absorbed by quality-related costs and where an internal audit should focus [S1][S3].
It is not adequate for product pricing, contract negotiation, line-specific investment approval or plant-to-plant performance ranking without normalization. Those decisions require actual annual dollars, product complexity, revenue definitions and failure data. A percentage benchmark alone cannot distinguish a plant producing $10 million from one producing $100 million, and it does not separate COPQ from the full cost of quality [S4][S6].
The method also needs a defined revenue basis. Sales revenue, production value and gross sales are not interchangeable in every calculation, yet each can appear in published cost estimates. Management should state the denominator, period, scope and treatment of appraisal costs before comparing periods or sites [S2][S4].
Reduction Priorities and Verification
The financial priority is to reduce the largest traceable failure categories, not to cut every quality expense. Start with scrap, rework, quality-stop downtime and warranty because each has a measurable source system. Then test whether prevention or appraisal spending can move downstream failures upstream without increasing escapes, false rejections or total inspection burden [S1][S5].
The 1-10-100 cost progression supports station-level detection and process correction, while the 500,000-unit example shows why scrap and rework rates need to be evaluated together [S2][S3]. Final-test data can also reveal which upstream process and supplier causes are driving defects, as detailed in EOL Test Data Analytics: Turning the Final Quality Gate into a Yield Engine.
Two signals justify a formal reduction program: COPQ remains above the plant's approved benchmark for two consecutive measurement periods, or one failure category such as downtime exceeds its own cost limit. A credible program should publish the annual COPQ dollar amount, COPQ per $1,000 of revenue, failure-category mix and avoided annual cost, then verify that the denominator remains unchanged [S4][S5].