Pareto Chart: Definition, 80/20 Rule, and How to Build One

July 11, 2026 · QualityManager Team

A Pareto chart is a bar chart ranking causes or defect types from most to least frequent, overlaid with a cumulative percentage line, showing which few categories account for most of the problem. It applies the 80/20 rule to quality data so teams see, at a glance, where fixing a handful of issues delivers most of the improvement.

What is a Pareto chart, exactly?

A Pareto chart combines two things on one graph: descending bars showing how much each category contributes (count, cost, or time), and a line showing the running cumulative percentage as you move left to right across the bars. The bars answer “what’s the biggest problem?” The line answers “how many categories do I need to fix before I’ve addressed most of it?”

The chart is named after economist Vilfredo Pareto, who observed that 80% of Italy’s land was owned by 20% of the population. Quality pioneer Joseph Juran later applied the same pattern to defects, calling it the distinction between the “vital few” and the “trivial many.”

A Pareto chart differs from a plain sorted bar chart in one important way: the cumulative line. Without it, you can still see which category is largest, but you can’t tell how many categories you’d need to address to clear, say, 80% of the problem. The line turns a ranking into a prioritization tool — it tells you where to stop investigating and start fixing.

What is the 80/20 principle behind Pareto analysis?

The Pareto principle (also called the 80/20 rule) states that roughly 80% of outcomes come from roughly 20% of causes. In manufacturing quality, that usually shows up as: a small number of defect categories generate most of your total defect count, scrap cost, or customer complaints.

The 80/20 split is a rule of thumb, not a law — real data might show a 70/30 or 90/10 pattern. What matters is the shape: a few categories dominate, and a long tail of minor categories contributes little. Pareto analysis exists to find that shape and act on it, which is why it’s a standard tool inside DMAIC’s Analyze phase (see our DMAIC guide) and pairs naturally with root cause analysis tools like the fishbone diagram once you know which category to dig into.

How do you build a Pareto chart from raw defect data?

Start with raw inspection or defect-log data over a defined period. Here’s a realistic example: a month of final inspection at a mid-size assembly line, 250 total defects logged across six categories.

Step 1: Tally raw counts by category.

Step 2: Sort descending by count, then add cumulative count and cumulative percent.

Cumulative % is just cumulative count divided by the grand total (250), expressed as a percentage. Each row adds the current category’s count to the running total from the row above.

Step 3: Draw the chart. Bars, in the sorted order above, show each category’s count on the left (primary) axis. A line plots cumulative percent on a right-hand (secondary) axis scaled 0-100%, with a point above each bar. The line starts at the height of the first bar’s percentage and climbs toward 100% as it crosses each subsequent bar.

Step 4: Read the vital few. Draw a reference line at 80% on the cumulative axis. In this data, scratched housing, label misalignment, and solder void together reach 78.8% — just short of three categories crossing the 80% mark, with missing fastener pushing it to 90%. That’s your vital few: four of six categories (67% of category types) account for 90% of all defects. The remaining two categories — wrong component and connector misseat — are the “trivial many,” contributing only 10% combined.

Reading the chart is mostly about the gap between bars, not just their absolute height. Notice the drop from 95 (scratched housing) to 62 (label misalignment) to 40 (solder void) — a steep descent, which is the visual signature of a genuine Pareto pattern. If the bars had instead come in at 45, 43, 41, 40, 39, 38, the chart would still sort fine, but there would be no vital few to chase, which is itself a useful (if less exciting) finding.

If you’d rather skip the spreadsheet setup, you can paste a defect table like the one above into QualityManager.AI’s free Pareto tool and it sorts the categories, builds the cumulative line, and highlights the vital-few cutoff automatically — useful when you’re re-running the analysis every month and don’t want to rebuild the chart by hand each time.

How do you act on a Pareto chart?

Once the vital few are identified, prioritize root cause investigation there first — not across all six categories equally. For the example above, that means:

1. Assign scratched housing first. At 38% of all defects, even a partial fix (say, cutting it in half) removes ~19% of total defects — more than eliminating any other single category entirely.
2. Run a fishbone or 5 Whys on the top 2-3 categories, not the whole list. Use a fishbone diagram template to branch scratched housing into machine, method, material, and manpower causes.
3. Re-pull the Pareto chart after corrective action. If scratched housing drops from 95 to 30 next month, the ranking shifts and a previously minor category may become the new priority — Pareto analysis is a repeat exercise, not a one-time chart.
4. Don’t ignore the trivial many entirely — track them, but don’t burn scarce investigation time there while the vital few remain unresolved.

A common mistake is treating the Pareto chart as the deliverable rather than the starting point. The chart tells you where to look; it doesn’t tell you why scratched housing happens 95 times a month. That’s a separate investigation — pull the parts, walk the line, interview the operators, and use a structured cause-and-effect method once the chart has pointed you at the right category. Skipping straight from “biggest bar” to “corrective action” without root cause work is how teams end up fixing symptoms instead of causes.

When does a Pareto chart mislead you?

Pareto charts fail in three common situations. Recognizing them prevents chasing the wrong priority.

Near-equal frequencies. If six categories each sit within a few percentage points of each other — no clear “vital few” emerges, and ranking them is closer to noise than signal. When bars are roughly the same height, the 80/20 pattern doesn’t apply, and you likely need to look at cost or severity instead of raw count, or reconsider whether your categories are defined too broadly (or too narrowly).

Counting occurrences when cost matters more. Raw count treats every defect as equally important, which is often wrong. Take the same 250-defect dataset above, but re-rank by cost impact — average rework or scrap cost per occurrence:

Ranked by cost, missing fastener — only the 4th-largest category by count — becomes the top priority, because each occurrence requires a full disassembly and part replacement. Scratched housing, the count-based #1, drops to third by cost because it’s usually a low-cost cosmetic touch-up. If your team had only looked at the count-based chart, they’d have spent the month on the least expensive problem. When severity, rework hours, safety risk, or scrap value vary significantly across categories, weight the chart by that measure instead of raw count.

Too-short data windows. A single shift or single day of defects reflects whatever happened to go wrong that day — a bad batch of components, an untrained temp operator, a machine that needed calibration — not your process’s real failure distribution. Pulling a Pareto chart from one day’s data can point you at a category that’s actually a one-off anomaly. Use at least four to six weeks of data, or enough total units to represent your normal production mix, before trusting the ranking.

A Pareto chart only earns trust when it’s built on a large enough sample, weighted by whatever actually costs you money or risk, and re-run regularly as corrective actions take effect.