Most cost-of-poor-quality explainers tell you that hidden costs exist and then stop, without ever showing the arithmetic that turns a defect count into a dollar figure. For a textile plant, that arithmetic starts with a number almost every quality team already tracks: DHU, defects per hundred units. This guide walks through turning that one familiar metric into an actual COPQ percentage, line by line. Walk through this calculation with iFactory AI on your own DHU data instead of an illustrative example.
Turning DHU Into an Actual Cost Number
Most COPQ explainers assert that hidden costs exist. This one shows the arithmetic: how defects-per-hundred-units data a textile quality team already collects becomes a real cost of poor quality percentage.
Why DHU Is the Right Starting Point
DHU, defects per hundred units, is one of the few textile quality metrics that nearly every plant already calculates, because it is simple: count the defects found in an inspected sample, divide by the number of pieces inspected, multiply by 100.
The problem is not that DHU is wrong. The problem is that DHU stops at a defect count, and a defect count alone cannot be put on a cost sheet. Turning it into cost of poor quality means attaching a labor minute, a material value, and a sorting hour to every defect that DHU already counted, rather than inventing a separate cost-tracking system from scratch.
This is also why most generic COPQ explainers leave a reader without an actual next step. They correctly point out that scrap percentages hide real cost, but they stop at the observation rather than the method, which leaves a quality team with a true statement and no way to act on it. Starting from DHU instead of from an abstract cost category is what makes this calculation something a plant can actually run this month, using data it already has sitting in its existing inspection reports.
The Four Numbers That Turn DHU Into Cost
A textile COPQ figure built from DHU data needs four inputs, each one answerable from records most plants already keep in some form.
Defects Per Day
Taken directly from the DHU count: defects found divided by pieces inspected, multiplied by 100, scaled to the day's total production volume.
Repair Time Per Defect
The minutes it takes to remove faulty stitching and redo it, specific to the defect type, since a skipped stitch and an open seam take different time to fix.
Daily Wage Rate
The labor cost per worker-day, used to convert repair minutes into a rupee or dollar figure rather than leaving it as a time estimate.
Rejection Value
The shipment price of pieces that cannot be repaired at all and are rejected outright, which is a separate cost category from repair time.
None of these four numbers requires new instrumentation to collect. Repair time per defect type is usually known informally by any experienced line supervisor, even if it has never been written down and multiplied against anything.
The reason these four specific numbers, rather than a longer list, are enough is that they map directly onto the three cost categories that actually show up on a plant's books: labor, material, and lost revenue. Repair time and daily wage rate together give labor cost. Rejection value gives lost revenue. A fourth category, appraisal cost, covered later in this guide, adds the quality team's own time. Four numbers, cleanly mapped to real cost categories, is a short enough list that a plant can actually maintain it going forward rather than abandoning it after the first calculation.
See This Calculation Run on Your Own Floor Data
Book a 30-minute session and see your DHU numbers converted into an actual cost of poor quality figure.
A Worked Example, Step by Step
The method is easier to trust once it is walked through with actual numbers, so here is one illustrative run using figures typical of a mid-size garment line, not a claimed result from a named plant.
| Step | Calculation | Illustrative Result |
|---|---|---|
| 1. Defects found | DHU 10 on stitching + DHU 15 on finishing, 1,000 pieces/day | 250 defects/day |
| 2. Repair time | 250 defects × 5 minutes each | 1,250 minutes ≈ 2.6 labor-days |
| 3. Repair cost | 2.6 labor-days × daily wage rate | A monthly repair-labor figure |
| 4. Rejection cost | Rejected pieces × shipment price per piece | A monthly rejection figure |
| 5. COPQ percentage | (Repair + rejection + appraisal cost) ÷ monthly shipment value × 100 | A comparable monthly COPQ % |
The number that comes out of step five matters less on its own than what happens when the same five steps are run again the following month, using the same wage rate and the same shipment-value basis. A single COPQ percentage is a snapshot. The same calculation repeated monthly is a trend, and the trend is what a quality program should actually be managed against.
It is worth noting what this worked example deliberately leaves out: a named plant, a specific dollar total, or a claimed percentage reduction. Those are exactly the kind of unsupported figures that show up in many generic COPQ pages without any stated method behind them. The value of walking through the five steps is that any quality team can rerun them with their own wage rate, their own repair times, and their own shipment prices, and arrive at a number they actually trust, because they watched it get built.
Where This Calculation Usually Breaks Down
The arithmetic is simple. Getting trustworthy inputs for it consistently is where most plants actually struggle, for a few predictable reasons.
Repair Time Is Guessed, Not Timed
A "5 minutes per defect" estimate that was accurate for one defect type gets applied uniformly to all of them, flattening real cost differences between defect categories.
Rejection Value Uses the Wrong Price
Rejected pieces are sometimes valued at cost rather than at lost shipment price, which understates the actual revenue impact of a rejection.
The Shipment-Value Basis Shifts
Comparing one month's COPQ percentage against another only works if both months use the same denominator; a changing order mix can quietly distort the trend.
Each of these three failure points has the same underlying cause: the calculation is run from memory or from a spreadsheet nobody updates consistently, rather than from defect records tied to an actual timestamp and an actual shipment value. The formula does not need to be more complicated to fix this. The inputs need to be more consistently measured.
What Appraisal Cost Adds That Repair and Rejection Miss
Repair and rejection cost are the two categories most plants calculate first, because they map directly onto defects. Appraisal cost, the salary of the people whose job is to catch those defects, is the category most often left out.
Why It Gets Skipped
Quality team salaries are treated as a fixed headcount cost on the org chart, disconnected from the defect volume that justifies that headcount in the first place.
Why It Belongs In the Number
A plant with a larger quality-checking team because its defect rate is higher is spending appraisal cost directly caused by poor quality, not an unrelated fixed overhead.
How To Add It Simply
Sum the monthly salaries of fabric, cutting, sewing, and finishing checkers plus the quality head, and add that single total into the COPQ calculation alongside repair and rejection.
Including appraisal cost is usually what moves a plant's COPQ percentage from a number that looks manageable to one that actually reflects what quality failures are costing across the whole operation, not just the visible repair and rejection line items.
A Short Self-Check Before You Calculate Your Own Number
Run these against your current quality tracking. A no on any of these points at where your COPQ figure, once calculated, will be least trustworthy.
Where iFactory AI Fits
iFactory AI captures defect counts, repair time, and rejection value automatically at the point they occur, and rolls them into a live COPQ percentage instead of a once-a-quarter manual calculation.
Automatic Defect Capture
DHU data is captured at the station where a defect occurs, tagged with defect type, so repair time can be matched to the actual category rather than a flat estimate.
Repair Time Tracking
Actual time spent on rework is logged per defect type over time, replacing a one-time guessed estimate with a continuously updated real figure.
Rejection Value Linkage
Rejected pieces are valued against actual shipment price per style, not a generic production-cost placeholder.
Rolled-Up COPQ Trend
Repair, rejection, and appraisal cost combine into one tracked percentage, updated continuously rather than recalculated by hand each quarter.
Expert Review
"A COPQ percentage is only as trustworthy as its weakest input, and in practice that weakest input is almost always repair time, because it gets estimated once and then reused for years regardless of which defect type is actually being repaired. A plant that times repair labor by defect category, even roughly, usually finds its real cost of poor quality is both higher and more unevenly distributed across defect types than its flat estimate ever suggested."
This review point is the practical takeaway worth acting on first: before trusting any COPQ percentage, check whether its repair-time input was ever actually measured or only ever assumed. Connect with iFactory AI's team to see repair time measured by defect type on your own line.
Frequently Asked Questions
We already track DHU. Why isn't that the same as tracking COPQ?
DHU counts how many defects occurred. It says nothing about how much those defects cost in repair labor, rejected shipment value, or the quality team's time spent catching them. COPQ is DHU with those three cost layers added on top, converted into a single comparable percentage rather than left as a defect count.
What is the actual formula for calculating textile COPQ?
Add repair labor cost, rejection cost (rejected pieces valued at shipment price), excess-material cost from re-cutting, and quality-team appraisal cost. Divide that total by monthly shipment value and multiply by 100. The resulting percentage is comparable month over month as long as the same wage rate and shipment-value basis are used consistently. Get a walkthrough to see this run on your own numbers.
Is there an industry-standard COPQ percentage we should be targeting?
No reliable published benchmark exists across the textile industry, and any specific percentage figure quoted online should be treated as one plant's illustrative result, not a target. The more useful goal is establishing your own baseline and tracking whether it moves in the right direction month over month.
Why does repair time per defect type matter more than a single flat estimate?
A skipped stitch and an open seam take different time to remove and redo, and applying one flat repair-time estimate to both flattens that real difference, understating the cost of the more time-consuming defect types and overstating the cheaper ones. Measuring repair time by defect category, even approximately, produces a materially more accurate COPQ figure.
What data do we need before calculating our first real COPQ number?
Daily defect counts by type (your existing DHU data), an approximate repair time per defect type, your daily wage rate, and the shipment price per style for rejected pieces. Most plants already have three of these four; the one usually missing is repair time broken out by defect type rather than as one blended estimate. Reach out to iFactory AI's team to scope what's already on hand.
Turn Your DHU Data Into a Real Cost Number
iFactory AI captures defects, repair time, and rejection value automatically and rolls them into a continuously updated COPQ trend. Book a walkthrough on your own floor data.
