Defect Rate Calculator
Count the defects, count the units inspected, and this calculator turns the pair into the numbers quality teams actually report: a defect rate percentage, its PPM equivalent, a DPMO with a sigma level when units can fail in more than one way, and a combined figure when several inspection lots need to roll up into one.
All three modes use the worked examples they are pre-filled with, so the figures in the Quick Answer below reproduce with a single click.
A process metric, not a verdict on any single batch. Compare against your own quality plan and sample sizes; small samples swing widely.
Quick Answer
Key takeaways
- Defect rate and PPM are the same number in different units: multiply the percentage by 10,000 to get PPM, so 2.40% and 24,000 PPM describe the identical result.
- Use the simple rate when each unit gets one pass or fail verdict. Use DPMO when a unit can fail in several independent ways, because DPMO divides by opportunities as well as units.
- The sigma scale is not linear: moving from 3 sigma (66,807 DPMO) to 6 sigma (3.4 DPMO) is roughly a 19,600 times reduction in defects, not a doubling of quality.
- Small samples lie. A rate built on 100 units can swing several percentage points either way, so judge a process on pooled, larger samples.
- Quick Answer
- Three modes, three different quality questions
- Defects, defective units, and opportunities: the distinction that changes the answer
- The formulas behind all three modes
- Worked examples, checked against this tool
- Where your rate sits: PPM targets and the sigma scale
- What competing calculators miss
- Frequently Asked Questions
- Add This Calculator to Your Website
- References
Three modes, three different quality questions
Each tab answers a different question, and picking the wrong one produces a number that looks precise and means the wrong thing.
Defect rate answers “what share of output went wrong?” when every unit gets a single pass or fail verdict. It returns the percentage and the PPM equivalent, and it refuses counts where defects exceed units, because that combination cannot describe pass or fail inspection.
DPMO + sigma answers “how capable is this process per chance to fail?” Enter defects, units, and the defect opportunities per unit. It returns DPMO, the sigma level with the standard 1.5 shift, defect-free yield, and a quality band from Poor to World class.
Multiple lots answers “what does the whole week, line, or supplier look like combined?” Add one row per lot and the tool totals units and defects before computing the combined rate, DPMO, and sigma, so a small clean lot cannot hide inside a big dirty one, or the other way round.
Defects, defective units, and opportunities: the distinction that changes the answer
A defect is one thing wrong. A defective unit is a unit with at least one defect. An opportunity is one distinct, checkable way a unit could fail.
A circuit board with 50 solder joints and two bad joints is one defective unit carrying two defects against 50 opportunities. Score it with the simple rate and you get one failure in one unit. Score it with DPMO and you get 2 defects in 50 opportunities, a far kinder and, for comparing board designs, far fairer figure. Neither number is wrong; they answer different questions, and mixing them in one report is how quality meetings go off the rails.
One naming trap deserves a direct warning. The similarly titled defect density calculator on this site does not compute a defects-per-unit figure: its tool measures a sales to net working capital finance ratio, and its article explains the software sense of defect density, defects per thousand lines of code, in prose. If you need defects per unit (DPU), divide total defects by total units yourself: the combined lots example below, 20 defects over 3,000 units, is a DPU of 0.0067.
The formulas behind all three modes
Three short formulas cover everything the tool computes, and the worked examples in the next section run each one.
The definitions line up with how practitioners use the terms: Minitab’s reference on DPMO describes it as the baseline metric for process capability precisely because it normalizes for complexity, and the NIST/SEMATECH engineering statistics handbook treats capability measurement the same way: a process is judged against its opportunities to fail, not just its unit count.
Worked examples, checked against this tool
Each mode is pre-filled with the example in its row, and the outputs below are what the tool returns for those exact inputs.
| Mode | Inputs | Outputs |
|---|---|---|
| Defect rate | 120 defects, 5,000 units (light bulb line) | 2.40% defect rate, 24,000 PPM, 97.60% first-pass yield |
| DPMO + sigma | 15 defects, 250,000 units, 5 opportunities per unit (door hardware) | 12 DPMO, 5.72 sigma, 99.9988% yield, band: Excellent |
| Multiple lots | Lot 1: 1,000 units, 12 defects, 5 opps. Lot 2: 2,000 units, 8 defects, 5 opps | 3,000 units, 20 defects, 0.67% combined rate, 1,333 DPMO, 4.50 sigma (Good) |
A second rate check, useful for software teams: 48 bugs found across 1,200 test cases is 48 / 1,200 = 4.00%, or 40,000 PPM. Enter it in the first tab and the tool returns exactly that.
Where your rate sits: PPM targets and the sigma scale
A rate means little until it is placed next to the numbers buyers and contracts use. The scale below is the standard Six Sigma conversion, including the 1.5 shift.
| Sigma level | DPMO | Defect rate | Reading |
|---|---|---|---|
| 6 sigma | 3.4 | 0.00034% | World class |
| 5 sigma | 233 | 0.0233% | Excellent |
| 4 sigma | 6,210 | 0.621% | Good |
| 3 sigma | 66,807 | 6.68% | Average |
| 2 sigma | 308,538 | 30.85% | Poor |
Industry targets sit well above the bottom of that scale: world-class automotive suppliers aim near 25 PPM, high-volume electronics near 500 PPM, and general manufacturing commonly runs 1,000 to 10,000 PPM. The worked example at 24,000 PPM sits between 3 and 4 sigma, respectable for an internal process and short of what an automotive contract would accept. One caution when reading the table: a defect rate in PPM equals DPMO only when each unit has exactly one opportunity. With 5 opportunities per unit, the same physical process shows a DPMO five times smaller than its PPM.
Sample size decides how seriously to take any of these placements. Two defects in 100 units reads as 2.00%, but the true rate behind it could plausibly be under 1% or over 6%. If you report rates to customers, pair the point estimate with a range: the confidence interval calculator turns the same counts into the plausible band around your percentage, which is the honest way to present a small-sample rate.
Defects also carry a cost beyond the percentage. Scrap, rework, inspection time, returns, and warranty claims are the cost of poor quality, a framework the ASQ cost of quality guide breaks into prevention, appraisal, and failure costs. When a rate will not move, costing what it destroys each month, for example with the cost of lost production calculator, is often what gets the fix funded.
What competing calculators miss
The defect rate tools ranking for this term split into two camps, and each camp leaves half the job undone.
Miss 1: the percentage with no context. Most simple tools return a percent and stop. A percent cannot be compared with a supplier contract written in PPM, or with a capability target written in sigma, so the reader is left doing unit conversions by hand. This page returns the percent and the PPM together, and its benchmark table places both on the sigma scale.
Miss 2: DPMO tools that cannot roll up lots. The DPMO calculators do the opportunities math well, one run at a time. Real inspection arrives in lots, shifts, and deliveries, and averaging several DPMO figures by eye weights a 100 unit lot the same as a 10,000 unit lot. The Multiple lots mode totals the raw counts first, which is the only correct way to combine them.
Miss 3: silent nonsense inputs. Several simple tools happily report a 150% defect rate when defects exceed units, a figure that signals a counting method problem, not a quality result. This page says so plainly and points you at the DPMO mode, which is where multi-defect counting belongs.
One honest limit: the sigma conversion assumes defect counts behave like a normal process with the conventional 1.5 shift. It is the industry-standard reporting convention, not a law of nature, and opportunity counts are always partly a judgment call. Two teams counting opportunities differently will produce different DPMO figures for the same line.
Frequently Asked Questions
It depends on the industry and the contract. Automotive suppliers often target around 25 PPM, high-volume electronics around 500 PPM, and general manufacturing commonly runs 1,000 to 10,000 PPM. Six Sigma’s benchmark is 3.4 DPMO. Compare your rate against your contract or your own history, not a universal number.
They are the same measurement in different units. Defect rate is per hundred units; PPM is per million units. Multiply a percentage by 10,000 to get PPM: 2.40% is 24,000 PPM.
DPMO is defects per million opportunities: defects divided by (units times opportunities per unit), times 1,000,000. Use it when one unit can fail in several independent ways, such as the 50 solder joints on one circuit board, because it fairly compares processes of different complexity.
Not in the simple rate mode, and not meaningfully: there the count is capped at the units inspected. If one unit carries several defects, the count can exceed the unit count, and the right measures are defects per unit or DPMO, which the DPMO mode computes.
First-pass yield is the share of units that pass inspection without rework. For a process with one pass or fail check per unit it is 100% minus the defect rate, so a 2.40% defect rate means a 97.60% first-pass yield.
More than feels intuitive. With 100 units and 2 defects, the underlying rate could plausibly sit anywhere from under 1% to over 6%. Track the rate over time, pool larger samples, and treat small differences between small samples as noise.
Sigma level = the standard normal quantile of (1 minus DPMO/1,000,000), plus 1.5 for the industry-standard long-term shift. A DPMO of 12 converts to about 5.72 sigma. The DPMO mode runs this conversion for you.
Lower is better. 3.4 DPMO is six sigma and world class, 233 is 5 sigma (excellent), 6,210 is 4 sigma (good), and 66,807 is 3 sigma (average). Most industrial processes aim for at least 3 to 4 sigma.
Yes. Use the Multiple lots mode: one row per lot with its units, defects, and opportunities per unit. The tool totals the lots and reports the combined defect rate, DPMO, and sigma level.
Add This Calculator to Your Website
References
- American Society for Quality (ASQ), “What Is Six Sigma?” – the quality profession’s overview of the Six Sigma method, its Motorola origins, and the long-term shift convention behind sigma levels. ASQ: What Is Six Sigma?. Accessed October 2026.
- Minitab, LLC, “What is DPMO?” – Minitab Statistical Software support reference defining defects per million opportunities as a baseline capability metric. Minitab Support: What is DPMO?. Accessed October 2026.
- NIST/SEMATECH, “e-Handbook of Statistical Methods: What is Process Capability?” – the federal engineering statistics handbook’s treatment of measuring a process against its specification and failure opportunities. NIST/SEMATECH e-Handbook: Process Capability. Accessed October 2026.
- NIST/SEMATECH, “e-Handbook of Statistical Methods” – handbook sections on statistical inference for proportions and rates, the basis for the small-sample cautions in this article. NIST/SEMATECH e-Handbook: Statistical Methods. Accessed October 2026.
- American Society for Quality (ASQ), “Cost of Quality (COQ)” – ASQ’s breakdown of prevention, appraisal, internal failure, and external failure costs used in the cost discussion above. ASQ: Cost of Quality. Accessed October 2026.