# Pareto Analysis

> Find the few causes behind most of a problem, and fix those first.

Chapter III, Problem solving · 3.10 · Origin: Vilfredo Pareto, applied to quality by Joseph M. Juran (1896) · Time: 15 to 30 min · Solo

**Use it when** many causes compete and you need the few that matter most.

Interactive workspace: https://kogu.tools/tools/pareto-analysis

## Abstract

Problems are rarely spread evenly across their causes. A handful of complaint types or customers usually account for most of the total. Pareto analysis shows this by counting each cause, sorting them from largest to smallest and keeping a running share.

If four categories out of twenty make up 80 percent of your complaints, that's where your time should go. The exact split varies from case to case, and the ranking is what you act on. The idea is named after Vilfredo Pareto, who noticed it in how land and wealth were distributed in Italy. Joseph Juran later applied it to quality problems.

## Method

1. **Choose what to measure.** Pick the thing you care about, like defects or cost, and the time period.
2. **Group into categories.** Sort the raw records into a manageable number of categories that don't overlap, so nothing gets counted twice.
3. **Count and sort.** Total each category and put them in order, largest first.
4. **Add the running share.** Work out each category's share of the total and the cumulative share as you go down the list.
5. **Focus on the top.** The categories that take you to about 80 percent are where to start.

## When to use it

- You have a long list of problems and need to decide where to start.
- You have counts or costs to work from, even rough ones.
- You need to show other people why one issue should come first.

## Common pitfalls

- Counting how often something happens when cost matters more. A rare problem can be the most expensive one.
- Categories that are too broad or overlap, which hide the pattern.
- Treating 80/20 as a law. The split varies, and the ranking is the part you use.
- Never coming back to the rest. Once the top causes are fixed, run it again.

## References

- Pareto, V. (1896). Cours d'économie politique. Lausanne: F. Rouge.
- Juran, J. M. (1951). Quality Control Handbook. McGraw-Hill.