The Cost of Poor Quality: How to Price a Single Defect

Cost of poor quality (COPQ) is a well-known idea in Six Sigma: sort quality-related spending into failure, appraisal, and prevention buckets, and you get a number leaders actually react to. What's harder is the step most teams skip past — pricing one specific defect, on one specific process, well enough that the number survives a skeptical finance review. A vague "this probably costs us a lot" doesn't open a project charter. A defensible per-occurrence figure does.

This is the practical half of that work: how to take a single, named defect and turn it into a number, step by step, using data you can actually get your hands on. For the underlying cost categories and vocabulary, our glossary entry on cost of poor quality (/blog/what-is-cost-of-poor-quality) covers the full picture; this post is about building one estimate for one defect.

What you're pricing, and what you're not

A department-wide COPQ estimate answers "how much does quality cost us, overall, this year?" That question matters for budgets, but it rarely drives a specific project. The question that drives a project charter is narrower: "how much does this one recurring defect cost us, and would fixing it pay for the effort?" That's a different exercise — smaller in scope, but it needs to be more precise, because someone is going to compare your number against the hours a project will actually consume.

Pick the defect first, before you price anything. "Quality problems in receiving" is not a defect; "wrong part number picked on an outbound order" is. A Pareto chart of defect codes or complaint reasons (/blog/root-cause-analysis-5-whys-vs-fishbone-vs-pareto) is usually how you get from the vague complaint to the specific, nameable defect worth pricing.

The four inputs that make up a defect's price

1. 01Detection cost per occurrence — the inspection, testing, or double-checking time already spent catching this specific defect before it moves downstream, priced at a fully-loaded labor rate.
2. 02Correction cost per occurrence — rework labor, scrapped material, and any re-testing needed once the defect is caught, whether it's caught internally or by the customer.
3. 03Escape cost per occurrence — the extra costs that only show up when the defect reaches the customer: expedited replacement shipping, credits or refunds, complaint-handling time, and the returns-processing labor involved.
4. 04Frequency over a defined period — how many times this defect actually occurred last month or last quarter, pulled from a defect log, a scrap tag count, or a returns report. Never substitute a guess here; frequency is the multiplier that turns a small per-occurrence number into a budget line worth acting on.

Multiply an occurrence's total cost (detection plus correction plus escape, when it applies) by the frequency, and annualize it. That single number — dollars per year attributable to this one defect — is what belongs in a project charter.

Where to find the numbers without a finance system

Most teams assume this data lives somewhere official and locked away. It usually doesn't. Labor rates come from payroll or HR, not a secret ledger. Time spent detecting or correcting a defect comes from a short time study — standing with a stopwatch, or asking the person who does the work to estimate their own time honestly, which is often more accurate than it sounds. Frequency comes from whatever log already exists: a scrap tag bin, a returns queue, a customer complaint tracker, a quality system's defect codes. If no log exists yet, running one for two to three weeks before pricing the defect is worth the delay — a frequency pulled from a real count beats a frequency pulled from memory every time.

Material cost is usually the easiest number to get and the smallest part of the total. The labor involved in catching and fixing the defect, plus the quieter cost of someone chasing down the correction, is typically the larger share — and it's the part teams forget to count because no line item on a standard report calls it "cost of this defect."

A worked example

Imagine a parts distributor tracking one defect: wrong SKU picked on an outbound order. A short time study shows that catching a mis-pick at the packing-verification step (detection) takes about four minutes of a packer's time per flagged order. When one gets through verification and ships wrong, correcting it means a return authorization, a reshipment, and a credit — call it forty-five minutes of combined warehouse and customer-service time, plus the cost of shipping the correct part overnight. A two-week count shows the defect occurring on roughly fifteen orders a week. Priced and annualized, the detection time alone is a modest number; the escaped cases, at forty-five minutes plus expedited freight each, dominate the total. The exercise also makes an invisible pattern visible: most of the annual cost comes from the smaller number of picks that escape verification entirely, not from the ones caught and corrected in house — which tells the project team exactly where to aim the fix.

The point of pricing a defect isn't to produce an impressive number. It's to produce a credible one that tells you which part of the process is actually worth fixing first.

Common ways this estimate goes wrong

• Treating a guessed frequency as data — go count occurrences for a real period before pricing anything
• Leaving out the labor of the person who catches or chases the defect, which is usually the largest cost and the easiest one to forget
• Inflating the number to make a project look more urgent — a conservative, defensible figure persuades a skeptical reviewer more than a dramatic one that doesn't hold up
• Confusing the cost of the defect with the cost of fixing it — these are two separate numbers, and a business case needs both to show a payback
• Pricing the defect once at kickoff and never revisiting it, which means the project can never prove how much it actually recovered

Where this fits in a project

This pricing exercise belongs in Define and Measure: it's what turns "this defect bothers us" into a charter a sponsor will actually fund, and it gives the team a baseline number to beat once Improve is underway. If the defect you're pricing is one of several candidates competing for attention, scoring them with an FMEA first helps you decide which one is worth pricing in this much detail, rather than running this exercise on every complaint that crosses your desk.

Building a full COPQ estimate and a defensible business case is core Green Belt territory — our Green Belt program (/courses/green-belt) walks through this pricing method as part of scoping a real DMAIC project, alongside the root-cause and measurement tools that tell you whether your fix actually worked. If you're earlier than that, our free White Belt (/courses/white-belt) introduces the vocabulary and the DMAIC structure this exercise sits inside, with the same timed, closed-book exam format and unlimited free retakes used across every level.