Cost of Quality and Cost of Poor Quality: The PAF Model and the Hidden Factory

The PAF model: prevention, appraisal, and failure costs

The Prevention-Appraisal-Failure (PAF) model, developed by Armand Feigenbaum in the 1950s and popularized by Philip Crosby and Joseph Juran, organizes quality costs into three categories that reflect the stage at which quality investment is made:

Prevention costs are expenditures made to prevent defects from occurring. They represent the quality investment made upstream, before production or service delivery, to design quality into the product and process. Prevention costs include: quality planning and design reviews, FMEA and Control Plan development, supplier qualification and development, statistical process control implementation and training, design of experiments for process optimization, process capability studies, and quality engineering support for new product launches. Prevention costs are the most leveraged category: one dollar spent on prevention typically avoids many dollars of appraisal and failure costs. However, prevention costs are also the most difficult to justify in the short term because their benefit is invisible — the defects that were prevented never appear in the data.

Appraisal costs are expenditures made to detect defects: to determine whether products and services conform to requirements before they are released to the next stage of the process or to the customer. Appraisal costs include: incoming inspection, in-process inspection and testing, final inspection and product release testing, calibration and maintenance of measurement equipment, laboratory testing, and quality audits conducted for conformance verification. Appraisal costs are necessary in any quality system that cannot rely entirely on prevention, but they are not value-adding: the inspector finding a defect does not add value to the product; the defect should not have been produced in the first place. A quality improvement strategy should aim to reduce appraisal costs over time by increasing process capability to the point where frequent inspection is no longer necessary.

Failure costs are the costs incurred when products or services fail to meet requirements. They are divided into internal failure costs (defects found before the product reaches the customer) and external failure costs (defects found after the product reaches the customer):

Internal failure costs include: scrap (the cost of material and labor in a part that cannot be reworked and must be discarded), rework (the labor and material cost of reprocessing a nonconforming part to make it conforming), repair (returning a product to a serviceable condition without meeting full conformance), reinspection and retesting after rework, downgrading (selling a product at a reduced price because it does not meet specification for its intended application), yield losses, machine downtime due to nonconforming material, and the cost of investigating and dispositioning nonconforming material (material review board time, corrective action investigation costs).

External failure costs include: warranty repairs and replacements, field service costs, product liability claims, recall and campaign costs, customer complaint handling, returned goods processing, loss of customer goodwill and future business, and regulatory fines or penalties associated with quality failures. External failure costs are almost always higher than the corresponding internal failure cost for the same defect, because the cost of finding and correcting a defect increases dramatically as it moves further down the value stream toward the customer.

The cost leverage of early detection

One of the most important principles in quality cost management is the rule of ten (also called the 1-10-100 rule): the cost of detecting and correcting a defect increases by roughly an order of magnitude at each stage of the value stream. A defect that costs one unit to detect and correct at the point of manufacture costs approximately ten units to detect and correct at final inspection, approximately one hundred units to detect and correct at the customer’s receiving dock, and approximately one thousand units to detect and correct in the field after the product has been put into service.

This cost leverage is the financial argument for investing in prevention and early detection. Moving quality controls upstream — from final inspection to in-process inspection, from in-process inspection to statistical process control, from SPC to mistake-proofing — reduces the average cost per defect by catching nonconformances earlier in the process. The same principle is the argument for investing in design reviews, DFMEA, and design-for-manufacturability work during product development: defects prevented or detected at the design stage cost a fraction of what they cost when discovered in production or in the field.

The hidden factory

The hidden factory is the portion of the production system devoted to finding and fixing quality problems rather than producing conforming output. It was named by Armand Feigenbaum to describe the invisible parallel production operation that exists in every factory: inspectors sorting bad parts from good, operators reworking nonconforming assemblies, engineers investigating scrap problems, material handlers moving rejected lots to quarantine areas, and managers in meetings discussing quality escapes. None of these activities add value to the product; they exist only because the primary production system is not running at sufficient quality levels to make them unnecessary.

The hidden factory is expensive for two reasons. First, it consumes real resources — labor, equipment time, floor space, and management attention — that could otherwise be used for productive output. Second, it is invisible to the conventional accounting system: scrap and rework are often recorded as overhead rather than attributed to specific quality failures, and indirect costs such as engineering investigation time, management meeting time, and customer service handling are rarely captured in quality cost reports at all.

Measuring and reporting quality costs

The starting point for a quality cost program is measurement: translating quality events into financial terms using the organization’s accounting data. Most quality cost data is already present in the organization’s financial system; the challenge is to extract and categorize it. Common data sources include:

Scrap and rework labor reports: the manufacturing execution system (MES) or ERP system typically tracks the labor and material cost of scrap and rework transactions. These data are the primary source of internal failure cost data.

Warranty and field service records: the warranty management system tracks claim counts, part replacement costs, dealer labor costs, and logistics costs. These are the primary source of external failure cost data.

Inspection and test labor: the number of inspectors, their labor cost, and the fraction of their time devoted to appraisal activities (as opposed to production support) constitutes the appraisal cost from the inspection function.

Engineering and quality support costs: the labor hours spent by quality engineers, process engineers, and managers on corrective action investigations, supplier quality activities, and customer complaint responses are prevention and failure costs that are rarely captured in formal quality cost systems but can be significant.

Quality costs are most useful when expressed as a ratio to a financial base (revenue, cost of goods sold, or labor hours) rather than in absolute dollars, because absolute figures make comparison across time periods and across different-sized plants meaningless. A quality cost report that shows total COPQ as a percentage of revenue, broken down by category (internal failure, external failure, appraisal, prevention), and trended over time provides the management insight needed to direct quality improvement investment.

Using CoQ data for quality improvement decisions

The purpose of a quality cost program is not just measurement but decision-making. CoQ data should drive three types of management decisions:

Investment prioritization: a Pareto analysis of internal failure costs (by product, process, defect type, or supplier) identifies the highest-cost quality problems and directs Six Sigma, lean, or corrective action resources to the problems with the highest financial return.

Prevention versus appraisal tradeoffs: when appraisal costs are high and stable while failure costs remain significant, it indicates that the inspection system is finding defects but not preventing them. This is the economic signal to shift investment from appraisal to prevention: process improvement, SPC, mistake-proofing, or supplier quality development.

Project justification: quality improvement projects can be justified in financial terms by quantifying the failure costs they will eliminate. A project that eliminates a specific scrap problem can be justified by the annualized scrap cost reduction. This financial framing makes quality projects directly comparable to capital investment proposals and operations improvement projects in the management budget process.

Common errors in quality cost management

• Measuring only the visible costs. Scrap and rework costs are relatively easy to measure because they appear directly in the manufacturing cost system. The hidden costs — engineering investigation time, customer complaint handling, premium freight to replace defective shipments, management meeting time on quality problems — are often two to three times larger than the visible scrap and rework costs but are invisible without a deliberate effort to capture them. A quality cost program that measures only scrap and rework understates COPQ by a large factor and underestimates the return on quality improvement investment accordingly.
• Using quality cost data punitively rather than constructively. If quality cost reports are used to assign blame or punish production managers for high scrap rates, the response will be to find ways to misclassify scrap as something else rather than to address the root cause. Quality costs are most effective as a management tool when they are used constructively: to identify improvement opportunities, to justify investment in prevention, and to measure the financial return of quality improvement projects.
• Failing to distinguish between prevention investment and failure cost. Some organizations classify all quality-related labor as “cost of poor quality,” including the work of quality engineers on prevention activities. This conflates investment in quality (a good thing to be increased) with the cost of failure (a bad thing to be reduced). A correctly structured quality cost report separates these categories so that the trade-off between prevention investment and failure cost reduction is visible.
• Not trending quality costs over time. A single quality cost snapshot has limited value. The power of quality cost data lies in trending: tracking how the mix of prevention, appraisal, and failure costs changes over time as quality improvement programs are implemented. A well-functioning quality system should show increasing prevention costs and decreasing failure costs over the improvement cycle.

Frequently asked questions

What is the difference between cost of quality and cost of poor quality?

Cost of quality (CoQ) is the total cost associated with quality: both the cost of achieving conformance and the cost of failing to achieve it. It encompasses all four PAF categories: prevention costs (investment in preventing defects), appraisal costs (investment in detecting defects), internal failure costs (the cost of defects found before the customer), and external failure costs (the cost of defects found after the customer). Cost of poor quality (COPQ) is the subset of CoQ that represents only the failure costs — internal and external — that arise when quality requirements are not met. COPQ does not include prevention or appraisal costs. The distinction matters because prevention and appraisal are investments in quality (made deliberately to achieve conformance), while failure costs are waste (the financial consequence of nonconformance that a better-performing quality system would eliminate). A quality improvement strategy aims to increase prevention investment and reduce failure costs, with the net effect that total CoQ decreases because the reduction in failure costs exceeds the increase in prevention investment. The term COPQ is often used loosely to mean all quality-related costs, but technically it refers only to the failure cost categories.

What is the rule of ten (or 1-10-100 rule) in quality cost management?

The rule of ten (also called the 1-10-100 rule) is the empirical observation that the cost of detecting and correcting a defect increases by roughly an order of magnitude at each stage of the value stream. A defect that costs one unit to detect and correct at the point where it is created costs approximately ten units to detect and correct at the next stage of the process (for example, final assembly inspection), approximately one hundred units to detect and correct when the product reaches the customer’s facility, and approximately one thousand units to detect and correct after the product has been put into service in the field. The cost increase at each stage is driven by: additional labor required to remove, disassemble, and replace the defective component (which was not necessary when the defect was caught at the point of manufacture); transportation and logistics costs; customer downtime and inconvenience; and the loss of customer goodwill that is difficult to quantify but is a real cost. The rule of ten is the financial argument for investing in upstream quality controls — process capability, SPC, mistake-proofing, and robust design — rather than relying on downstream inspection and warranty to catch defects.

What is the hidden factory?

The hidden factory, a concept named by Armand Feigenbaum, is the portion of an organization’s capacity devoted to finding and correcting quality problems rather than to producing conforming output. It is “hidden” because conventional accounting systems record these activities as overhead rather than attributing them to specific quality failures, making the true cost of poor quality invisible to management. The hidden factory includes: inspectors sorting nonconforming parts, operators reworking defective assemblies, quality engineers investigating scrap problems, purchasing teams expediting replacement material for defective shipments, customer service representatives handling complaints, logistics staff managing returns and replacements, and managers in meetings discussing quality escapes and corrective actions. In most manufacturing organizations, the hidden factory consumes 15 to 40 percent of total capacity. Making the hidden factory visible — by measuring and reporting the labor, floor space, equipment time, and overhead consumed by quality problem activities — is the first step toward eliminating it. A Six Sigma DMAIC project that eliminates a significant quality problem typically reduces the hidden factory by an amount that frees up real production capacity without capital investment.

How should quality costs be expressed for management reporting?

Quality costs are most useful for management reporting when expressed as a ratio to a financial base rather than in absolute dollar terms, because absolute figures are not comparable across time periods with different production volumes or across facilities of different sizes. Common normalization bases include: quality cost as a percentage of revenue (the most commonly used), quality cost as a percentage of cost of goods sold (COGS), and quality cost per labor hour. The choice of base affects the apparent trend: if revenue grows faster than quality costs, the quality cost ratio will fall even if absolute quality costs are unchanged. A quality cost report for management should show: total COPQ as a percentage of revenue, the breakdown by category (internal failure, external failure, appraisal, prevention), a time series showing the trend over the previous 12 to 24 months, and a Pareto of the major contributors to internal and external failure costs. This format gives management the financial scale of the quality problem, the trend direction, and the specific issues where investment in improvement will have the highest return.

What is Philip Crosby’s “quality is free” concept?

Philip Crosby’s “quality is free” concept, introduced in his 1979 book of the same name, argues that investing in quality improvement is not a cost but a financial benefit. The argument is based on the observation that the cost of implementing prevention programs (training, process improvement, mistake-proofing, SPC, quality planning) is almost always less than the cost of the failure costs those programs eliminate. In most organizations, COPQ is 15 to 40 percent of revenue; the investment required to reduce COPQ by half is typically a small fraction of that savings. Therefore, the net financial impact of quality improvement is positive: the organization saves more in reduced failure costs than it spends on prevention. “Quality is free” does not mean that quality costs nothing to achieve; it means that the investment pays for itself and generates a positive net return. Crosby’s framework was instrumental in shifting management’s perspective on quality investment from a cost to be minimized to an investment with a measurable financial return. His “zero defects” goal was the operational expression of the same principle: the only acceptable quality target is zero defects, because any non-zero defect target implicitly accepts a level of COPQ that could have been eliminated.

How are quality costs used to justify Six Sigma projects?

Six Sigma projects are typically justified using quality cost data by quantifying the financial value of the defect or process problem the project will address, then comparing that value to the cost of running the project. The most common approach is to identify a specific internal failure cost (a chronic scrap problem, a high-volume rework operation, a production line yield loss) or external failure cost (a warranty claim category, a customer return issue) and calculate its annualized cost from the organization’s quality cost records. The project is then justified by the projected reduction in that cost: if a project will eliminate 80 percent of a scrap problem that costs $500,000 per year, the project is projected to generate $400,000 per year in savings. This financial framing makes Six Sigma projects directly comparable to capital investment proposals in the management budget process. For external failure costs, the calculation also includes the premium freight, customer deduction, and potential lost business costs associated with the quality problem — which are often substantially larger than the direct part replacement cost. Organizations with mature quality cost systems can track the actual versus projected savings for completed projects, providing feedback to improve the accuracy of future project financial justifications.