Full Inspection, Increased Sampling, and Re-inspection, Yet Defects Persist — Five Steps to Diagnose Quality Cost Structure and Shift Investment from Inspection to Prevention

A monthly quality cost report for an automotive parts company sits on the general manager's desk: last year's average monthly quality cost was 850,000 yuan, and this year it has risen to 1,320,000 yuan. Flipping through the details, the most significant increase is in the inspection cost — the number of inspection personnel has increased from 12 to 21, inspection hours, testing materials, and external testing have all doubled, raising this category from 300,000 yuan to 610,000 yuan. By common logic, doubling the inspection investment should result in a cleaner interception of defects, but the lower half of the same report reads: the internal nonconforming rate has risen from 2.8% to 3.1%, and external failure costs (customer complaint handling, returns, and claims) have increased from 180,000 yuan to 270,000 yuan.

The general manager asked just one question: "Why are problems still increasing despite doubling the inspection budget?"

This is a question almost every manufacturing company will encounter. It's not about whether the quality investment is sufficient, but rather where the money is being spent. The real issue with quality costs often lies in the structure, not the total amount.

1. First, Understand One Thing: Inspection Buys "Interception," Not "Quality"

Quality costs (Cost of Quality, COQ) are typically classified into four categories according to international standards:

Many companies focus only on the total figures in the first and fourth columns, i.e., "How many million yuan is the quality cost this month?" However, breaking down the four categories and examining their proportions makes the issues clear.

The core principle is straightforward: Inspection is a "sieve," not a "maker." Appraisal costs buy information and interception capabilities — they tell you where the defective products are and how many have been intercepted, but they do not change the probability of defects being generated by the process. If a process inherently produces a 3% nonconforming rate, adding three more inspections downstream will still leave the process with a 3% nonconforming rate. You are just spending more manpower to pick out the same defects and shifting the losses from "external" to "internal." The only way to reduce that 3% is through prevention: poka-yoke, parameter control, standardized work, supplier improvement at the source, and process margin.

This is the classic 1-10-100 rule: a problem that costs 1 yuan to solve in prevention will cost 10 yuan to detect internally and 100 yuan to handle externally. If the balance is consistently skewed towards appraisal and failure, the company's quality costs will fall into a vicious cycle of "increased investment, unchanged defects" — you are spending money to address problems that have already occurred, rather than reducing their occurrence.

To determine if your structure is imbalanced, look for these three signals:

• The growth in inspection personnel or inspection hours is faster than the growth in production volume and value. This indicates that the frequency of problems has not decreased and can only be intercepted through a human sea strategy.
• The first reaction to a customer complaint is to "add another inspection." This is the most typical organizational inertia in an imbalanced structure.
• Prevention-related investments are consistently below 10% of the total quality cost. There are no budget items for training, poka-yoke, SPC, or supplier coaching, or these items are the first to be cut each year.

A relatively healthy cost structure typically has prevention costs accounting for 20% to 30%, appraisal costs 30% to 40%, and failure costs (internal + external) controlled within 30%. An imbalanced structure often looks like this: prevention costs below 10%, appraisal costs over 50%, and failure costs over 30% — the majority of the bill is for "in-process interception" and "post-event losses," with almost no "pre-event construction."

2. Five-Step Method: Diagnose the Structure and Shift Investment

Step One: Standardize the Calculation, Break Down the Four Categories to the Line or Product Family Level.

The prerequisite for structural diagnosis is that the data can be broken down. Calculating a total for the entire factory will never reveal where the money is being spent. This step involves three actions:

1. Define and Document the Calculation Rules. Financial terms do not include an "appraisal cost" category, so a mapping rule is needed: how to allocate inspection personnel hours (by proportion of hours or actual recorded hours), whether to classify testing equipment depreciation and calibration costs as appraisal or prevention (calibration costs are generally classified as appraisal, while annual capability studies and tool improvements can be classified as prevention), and whether re-inspection after rework is classified as internal failure or appraisal (it should be classified as internal failure to avoid double counting).
2. Uncover Hidden Items. The most easily overlooked items in the report are: production capacity losses due to line stoppage for judgment, air freight and excess freight for expedited replenishment, labor costs for customer complaint handling and on-site service, claims and penalties, price differences due to downgraded use, and additional full inspection hours for a single customer complaint. If these are not included, failure costs will be significantly underestimated.
3. List by Line, Product Family, or Project. Output a "Quality Cost Structure Ledger" with each unit on one line, the four categories of costs listed separately, and sales revenue annotated to facilitate ratio calculations.

Step Two: Calculate Three Structural Indices, Compare with the Baseline, Not Last Year.

First, the ratio of prevention costs to sales revenue; second, the ratio of appraisal costs to the combined total of appraisal and prevention costs (reflecting whether the investment is biased towards "screening"); third, the ratio of failure costs to the total quality cost. Viewing these three numbers together will immediately reveal structural imbalances.

The actual structure of a certain electronics assembly line (monthly sales of 12 million yuan) is as follows:

The information from this table is clear: every 1 yuan spent on prevention corresponds to 6.7 yuan in appraisal costs and 4.7 yuan in failure costs. The key issue is not that the quality cost as a percentage of sales revenue (3.33%) is high, but that 92% of this 3.33% is spent on post-event screening and post-event compensation.

Step Three: Identify "Inspection-Intensive" Processes, Lock in Priority Targets for Migration.

Break down the appraisal and failure costs by process and create a cross-tabulation: the horizontal axis is the percentage of inspection hours for each process, and the vertical axis is the percentage of rework, scrap, and line stoppage hours related to each process. Processes that fall into the high-high zone are the top priority for investment migration — these processes consume a large amount of inspection resources and continue to generate losses, indicating that their issues cannot be resolved through inspection alone and require process improvement.

Conversely, processes that fall into the "high inspection, low failure" zone are often critical characteristics (such as safety features or performance tests) that must be fully inspected due to product or regulatory requirements and should not be targeted for reduction. This is important: shifting investment does not mean indiscriminately cutting inspections.

Step Four: Shift the Budget from "Appraisal" to "Prevention," Item by Item.

After completing the structural diagnosis, specific migration actions must be taken; otherwise, it remains just a report. The logic of migration is: first, ask "What is this inspection intercepting," and then ask "Can this issue be prevented from occurring?" Below are common capability gaps and their corresponding migration paths.

Migration is not a one-time removal of inspections but a "build the defense first, then withdraw the old actions" approach: new methods should be implemented and run stably for a period to verify that the failure rate has indeed decreased before considering reducing the inspection frequency. Reversing this order is like dismantling the old defense before the new one is in place.

At the same time, preventive investments must be specific, budgeted projects with responsible persons and delivery times — whether it's a poka-yoke device, an SPC parameter, or a round of on-site supplier coaching. Otherwise, "strengthening prevention" will remain just a meeting slogan.

Step Five: Verify with Dual Indicators, Not Just Total Costs.

The effectiveness of migration cannot be judged solely by the total quality cost, as the total includes both intentionally reduced inspection costs and newly added preventive costs, which may offset each other in the short term. Four directions should be considered: an increase in the percentage of preventive costs, a decrease in appraisal costs, a decrease in failure costs, and a decrease in the nonconforming rate. It is recommended to review these at 3-month, 6-month, and 12-month intervals.

Case Study: A home appliance motor production line had a 100% electrical performance test that consistently occupied six inspection personnel, making the test station a bottleneck for the entire line. The diagnosis revealed that defects were concentrated in winding parameter deviations, which could be monitored during the winding process. The improvement plan was to invest 60,000 yuan in poka-yoke devices and parameter SPC at the winding process. After three months of stable operation of the poka-yoke and alarm systems, the 100% performance test at the end of the line was changed to batch sampling. The result was: inspection personnel reduced from six to three, the internal nonconforming rate dropped from 2.6% to 0.9%, and the monthly quality cost decreased from 410,000 yuan to 280,000 yuan — with preventive costs increasing from 15,000 yuan to 40,000 yuan, appraisal costs decreasing from 220,000 yuan to 110,000 yuan, and failure costs decreasing from 175,000 yuan to 130,000 yuan. Over 12 months, the cumulative savings exceeded 1.5 million yuan, and the main source of net gains was not the reduction in inspection personnel but the decrease in defects themselves.

3. Six Common Pitfalls

Pitfall One: Treating COQ as a Financial Statement, Calculating Without Using. The data may look impressive, but if no one uses it for decision-making, it is just a formality. The purpose of quality costs is to guide resource allocation, and the trend of each category should correspond to the budget actions for the next quarter. Otherwise, even the most accurate calculations are useless.

Pitfall Two: Cutting Appraisal Costs Without Increasing Preventive Investment. This is the most dangerous approach. Reducing inspections without establishing new defenses is equivalent to pushing problems directly to the customer, leading to a rapid rebound in external failure costs, often exceeding the savings from inspection cuts.

Pitfall Three: Using "Reducing Total Quality Costs" as the Sole Performance Metric. Under total cost metrics, the easiest to cut is preventive costs — because they "appear to produce nothing." The result is that the numbers look good in the first year, but failure costs double in the second year, returning to the starting point. The metrics must include a minimum threshold for the percentage of preventive investment.

Pitfall Four: Misunderstanding "Strengthening Prevention" as Repeated Emphasis in Meetings. If preventive investments are not specific projects with budgets, responsible persons, and timelines, they are just expressions of intent. Whether to purchase poka-yoke devices, implement SPC systems, or schedule supplier coaching must be listed in specific action items.

Pitfall Five: Inconsistent Calculation Rules for the Four Categories, Leading to Conflicting Numbers. The quality department and the finance department use different calculation methods, resulting in different numbers reported at two meetings. Management will quickly lose trust in this data. Once the rules are set, they should be documented and comparable across years.

Pitfall Six: Underestimating the Hidden Part of Failure Costs. Calculating only scrap and rework costs, without considering line stoppage, expedited freight, customer penalties, and trust loss, systematically underestimates the cost of failure. This is why many companies think "adding another inspection is cheap" — because their accounts do not reflect the true cost of failure.

4. In Summary

If quality costs cannot be reduced, it is often not because of insufficient spending but because of structural errors — shifting the budget from "post-event screening" to "pre-event prevention" will naturally lower the total cost.

The key to quality costs is structure, not total amount

Knowledge code: 4.3.1

Version: v20261008

Author: QTank QTank is dedicated to providing systematic professional knowledge, methodologies, and practical tools to quality management practitioners, helping companies continuously improve their quality capabilities.