Description
Ranks problem or error categories by frequency in a bar chart with a cumulative percentage line, to find the few causes behind most problems.
A Pareto diagram orders the frequencies of defect or error categories from most to least frequent, with a cumulative percentage line, to separate the vital few from the trivial many. Its rigour lies less in the chart than in the data behind it. In the worked example a health insurer aiming to halve claim-processing errors forms a cross-functional team, agrees a list of error categories from experience and a process review, audits a random sample of 1,000 claims end to end, and records every error by category, with agreed coding rules for ambiguous cases and for cases carrying several errors. Frequencies and cumulative percentages are tabulated and the leading categories become improvement targets. The chart also communicates priorities clearly to management and production teams.
Attributes
- Discipline
- Product Analytics & Strategy
- Loss type
- Siloed
- Goal
- Prioritise, Diagnose errors
- Stage
- Evaluation, Post-launch
- Data
- Quantitative
- Effort
- Days
Steps
- Form a team spanning the stages of the process or journey under study.
- Agree a list of problem or error categories from experience and a process review.
- Take a random sample of cases (for example 1,000 transactions or sessions) and audit each end to end.
- Record every problem by category, agreeing rules for ambiguous and multiple codings and checking coder agreement.
- Tabulate category frequencies and cumulative percentages, and draw bars in descending order with the cumulative line.
- Focus improvement on the few categories that account for most problems, then repeat after changes to confirm the shift.
Sources
- Statistics for Business and Economics (Newbold et al., 2022)
Also known as
- Pareto diagram analysis
- Pareto chart
- 80-20 rule analysis
- Vital few and trivial many
Categories
- Analytics & Behaviour
- Product Analytics & Strategy