Make your chart
What is a Pareto chart?
A Pareto chart is a bar chart sorted from the largest category to the smallest, with a line showing the cumulative percentage of the total. It is named after the Pareto principle, the observation that a small share of causes often accounts for a large share of the effect. In quality work this is often stated as "80/20", but the split is rarely exactly 80/20. The chart exists to show your real split, not to assume one.
How to read it
- Bars show the count for each category, largest on the left.
- The cumulative line shows what share of the total the categories up to that point explain.
- The dashed 80% line marks the common cutoff. Categories up to and including the one where the cumulative line first reaches 80% are the vital few; the rest are the "useful many".
- A steep curve that reaches 80% within two or three bars means focus will pay off. A flat curve (bars of similar height) means there is no dominant cause, and you may need to re-stratify the data a different way.
Worked example (example data)
The sample loaded above is made-up example data for a delivery team that logged 100 customer complaints in a month:
| Category | Count | Cumulative % |
|---|---|---|
| Late pickup | 42 | 42% |
| Wrong address | 27 | 69% |
| Damaged in transit | 14 | 83% |
| Missing items | 9 | 92% |
| Billing error | 5 | 97% |
| Other | 3 | 100% |
The cumulative line crosses 80% at the third bar, so three of six categories (late pickup, wrong address, damaged in transit) explain 83% of complaints. The team would start root-cause work on late pickups, because that one category is 42% of the problem by itself.
Common mistakes
- A large "Other" bar. If "Other" is one of your biggest bars, your categories are too broad. Break it down before drawing conclusions. (This tool keeps "Other" where its count puts it; many practitioners move it to the far right.)
- Counting frequency when cost matters. Ten cheap defects may matter less than two expensive ones. Make a second chart weighted by cost or time when the impact per item differs.
- Mixing time periods or processes. Use one consistent period and scope, or the ranking may reflect the mix rather than the process.
- Too little data. With only a handful of occurrences, the order can flip from week to week. Collect enough data for the ranking to be stable.
- Treating 80% as a law. It is a guide. If the top two bars explain 65%, they may still be the right place to start.
- Overlapping categories. Each occurrence should fall into exactly one category, or the totals double-count.
When to use a Pareto chart in DMAIC
Measure phase: once you are collecting defect or complaint data with a clear operational definition, a Pareto chart shows where the volume sits and helps narrow the project scope to the categories that matter.
Analyze phase: drill down. Take the top bar and draw a second-level Pareto (for example, late pickups by route, shift or customer type) to point root-cause tools like fishbone diagrams and 5 Whys at the right place. After the Improve phase, a before-and-after Pareto on the same categories is a simple way to show the change.
Privacy
All calculations and the chart are done with JavaScript on this page. Nothing you type is sent to a server or stored.
Want it in a spreadsheet?
This free tool comes first and will stay free. If you would like the same analysis in a workbook you can keep and reuse, there is an optional Pareto Analysis Calculator template (), or the Everything Bundle (9) with all eight templates. See the template store for details.
Made by a Six Sigma Master Black Belt with 17 years of Lean Six Sigma process improvement experience.