Pareto analysis
Most of the effect comes from a few of the causes. Sorts categories by frequency from largest to smallest and draws a cumulative percentage curve on top, so the handful of items behind most of the problem becomes visible.
Register nowWhen to use it?
- Deciding which of many complaints, defects or cost items to tackle first.
- Showing where an improvement will pay off most.
- Arguing for concentrating resources on a few points rather than spreading them.
Step by step
- 1Create the analysis on first open. Use the list at the top to switch between analyses in the project, the pencil to rename, the bin to delete.
- 2Enter the category name and its frequency in the table on the left panel.
- 3Use the add button below the table for a new row.
- 4The chart updates instantly: bars sort from largest to smallest and the curve shows the cumulative percentage.
Keyboard shortcuts (desktop)
Tips
- You can enter cost or lost time instead of frequency, as long as every row uses the same unit.
- Stop where the curve flattens: the long tail on the right is the part not worth chasing.
- Slice the categories too finely and nothing stands out — the chart just flattens. Merge similar items.
Worked example
Example: where the customer complaints come from
An online shop received 480 complaints in three months. The team had been discussing a separate fix for each complaint type. Counting them and sorting largest first changes the conversation.
Complaint type and count
- Delivery was late — 196
- Item did not match the description — 121
- Returns process too slow — 62
- Item arrived damaged — 48
- Wrong item sent — 29
- Other — 24
Cumulative share
- Late delivery — 41%
- + Description mismatch — 66%
- + Slow returns — 79%
- + Damaged in transit — 89%
- Remaining three — 100%
The top two account for two-thirds of everything. Instead of chasing six problems at once, fixing delivery times and product descriptions removes 66% of the dissatisfaction. Making that order visible is the entire job of Pareto analysis.
Frequently asked questions
What is Pareto analysis?
A method that sorts problems by how often they occur, largest first, and shows which few account for most of the total. It rests on a simple observation: roughly 80% of the effects come from roughly 20% of the causes.
Does the 80/20 rule always hold?
Not exactly, and it does not need to. Sometimes it comes out 70/30, sometimes 90/10. What matters is not the ratio but that the distribution is uneven: if a few items carry most of the total, Pareto analysis is useful.
Should I sort by count or by cost?
By whatever your decision depends on. Count shows which problem happens most; cost shows which one hurts most. They often disagree — a rare but expensive problem sits at the bottom of a count-based list.
How many categories should a Pareto chart have?
Five to ten reads best. An analysis with thirty categories is still just a list and gives you no focus. Choosing a small number of genuinely distinct categories is half the work.
Is it free?
Yes. Klarsti is free and ad-free right now, and you do not need an account to run a Pareto analysis.