A quality team may complete hundreds of audits every month and still struggle to answer one important question:
What should we fix first?
The monthly report might show:
Overall quality: 93%
Total audits: 600
Total errors: 200
Critical errors: 4
These numbers describe performance, but they do not automatically identify the best improvement opportunity.
To make the data actionable, management needs to understand which error categories account for the largest share of observed defects.
That is where Pareto analysis becomes useful.
What Is Pareto Analysis?
Pareto analysis is a method of ranking problems by their frequency, cost, impact or another meaningful measure.
It is commonly associated with the 80/20 principle: a relatively small number of causes may account for a large share of the consequences.
In quality management, the idea is simple:
Identify the error categories contributing most to the problem, then investigate those categories first.
The 80/20 relationship is a useful heuristic, not a mathematical law. Your data might show that three categories create 76% of defects, or that five categories create 62%.
The objective is not to force the result to equal 80%.
The objective is to identify where improvement effort is likely to have the greatest value.
A Practical QA Pareto Example
Suppose a back-office quality team records 200 defects during the month.
The findings are grouped into six error categories:
| Error category | Error count | Share of defects | Cumulative share |
|---|
| Incorrect validation | 72 | 36% | 36% |
| Missing documentation | 48 | 24% | 60% |
| Incorrect processing | 32 | 16% | 76% |
| Incorrect calculation | 24 | 12% | 88% |
| Communication error | 16 | 8% | 96% |
| Other | 8 | 4% | 100% |
| Total | 200 | 100% | |
The first three categories account for:
72 + 48 + 32 = 152 defects
That is:
152 ÷ 200 × 100 = 76%
Management now has a much clearer starting point.
Instead of saying:
“We need to improve quality across the team.”
the discussion becomes:
“Three error categories account for 76% of our observed defects. Let's investigate why those errors are occurring.”
That is the practical value of Pareto analysis.
How to Build a Pareto Chart
A Pareto chart combines two views.
Bars show the number of defects in each category, arranged from highest to lowest.
A cumulative line shows how much of the total is explained as each category is added.
For the example above, the cumulative percentages are:
36% → 60% → 76% → 88% → 96% → 100%
The chart makes it easy to see where the largest concentration of defects sits.
It also helps management avoid spending equal effort on every category when the data suggests that some deserve more attention.
Step 1: Define What You Are Counting
Before building the chart, decide what the measure represents.
For example:
Number of defects
counts individual errors.
Number of defective audits
counts audits containing at least one error.
Number of critical-error audits
counts audits containing at least one critical error.
These are different measures.
One transaction may contain three defects. If you are counting defects, it contributes three to the total. If you are counting defective transactions, it contributes one.
The Pareto chart should clearly state which measure is being used.
Step 2: Use Consistent Error Categories
Pareto analysis depends on good error classification.
Categories such as:
Incorrect processing
Processing issue
Wrong process
may overlap and make the results difficult to interpret.
A stronger taxonomy uses clear definitions and avoids unnecessary duplication.
For example:
Category: Validation
Error type: Required validation not completed
Category: Documentation
Error type: Mandatory evidence missing
Consistent categories make trends more reliable and help managers compare one period with another.
Step 3: Count and Rank the Errors
Once the audit findings are classified, count the number of defects in each category.
Then sort them from highest to lowest.
The largest category appears first.
This ranking provides the basic Pareto view.
However, the largest category should not automatically become the first corrective-action project. Management still needs to consider severity, risk and whether the problem is realistically addressable.
Step 4: Calculate the Cumulative Percentage
For each category:
Cumulative Percentage = Cumulative Error Count ÷ Total Error Count × 100
Using the example:
First category:
72 ÷ 200 = 36%
First two categories:
(72 + 48) ÷ 200 = 60%
First three categories:
(72 + 48 + 32) ÷ 200 = 76%
This shows how quickly the largest categories account for the total defect population.
Pareto Analysis Should Consider Severity
Frequency is important, but it is not the same as risk.
Imagine a quality report showing:
250 formatting errors
18 incorrect customer decisions
3 serious privacy failures
A frequency-based Pareto chart will emphasize formatting.
That may be appropriate if the objective is to reduce the total number of defects.
But the three privacy failures may deserve more urgent attention because of their potential impact.
A mature quality program should therefore consider both:
How often does the error occur?
and:
How serious is the error?
For more on this distinction, read Critical vs Major vs Minor Errors: How to Design a Better QA Error-Severity Framework.
Use More Than One Pareto View When Necessary
Different management questions may require different Pareto analyses.
Pareto by Error Count
Which error categories occur most frequently?
Pareto by Critical Errors
Which categories account for the greatest number of critical findings?
Pareto by Worktype
Which errors are most common within a particular process?
Pareto by Team
Which categories are driving quality loss for a particular team?
Pareto by Rework
Which errors generate the most additional processing effort?
These views should not be blended into one unexplained score.
Each should answer a specific question.
The Largest Error Category Is Not Always the Highest Error Rate
Consider two Worktypes.
Worktype A
1,200 audits
24 defective audits
Defective-audit rate: 2%
Worktype B
80 audits
12 defective audits
Defective-audit rate: 15%
Worktype A has more defective audits in absolute terms.
But Worktype B has a much higher defect rate.
If management looks only at counts, Worktype A may appear to be the bigger problem.
If management looks only at rates, it may overlook the larger total number of defects in Worktype A.
Both views matter.
Counts show the volume of the problem. Rates show the problem relative to the exposure.
Be Careful with Targeted Sampling
Suppose the QA team intentionally increases audits for a high-risk Worktype.
That Worktype may now contribute a larger share of observed defects simply because it received more audit attention.
This does not necessarily mean its underlying quality deteriorated.
When interpreting Pareto results, ask:
Was the sample random or targeted?
Did sampling change this month?
Did one Worktype receive more audits than usual?
Are we comparing similar populations?
For more detail, read QA Sampling: How Much Should You Audit and How Should You Select the Sample?.
Pareto Tells You Where to Investigate—Not Why the Error Happened
This is one of the most important limitations.
Suppose incorrect validation accounts for 36% of defects.
Pareto analysis tells management:
Incorrect validation is the largest observed error category.
It does not prove the root cause.
Possible causes might include:
The next step is investigation.
A useful sequence is:
Pareto → Root-Cause Analysis → Corrective Action → Follow-Up Measurement
Without root-cause analysis, an organization may simply tell employees to “be more careful” and see little improvement.
Use Pareto Analysis to Make Coaching More Specific
Suppose an employee has ten quality findings.
Seven relate to the same validation step.
Instead of providing ten separate pieces of generic feedback, the supervisor can identify the recurring pattern.
The coaching discussion becomes:
“Most of your recent errors relate to this validation requirement. Let's review that step and then check whether the issue improves.”
This is more useful than focusing only on the employee's overall score.
It also connects naturally to a closed-loop feedback process.
Read From QA Finding to Improvement: How to Build a Closed-Loop Quality Feedback Process.
Pareto Can Reveal Process-Wide Problems
Suppose one employee repeatedly makes a particular error.
That may suggest an individual coaching need.
Now suppose the same error appears across 40 employees after a procedure change.
The likely improvement opportunity may be broader.
Management should investigate whether:
Pareto analysis can therefore help distinguish between a concentrated individual issue and a wider process problem.
Turn the Pareto Result into an Improvement Plan
The purpose of the chart is not to create another dashboard.
It is to support action.
For example:
Finding: Incorrect validation is the largest error category.
Investigation: Review audit evidence and identify the most common validation failures.
Root cause: Determine whether the issue relates to training, procedure, system design or another factor.
Action: Introduce the appropriate corrective measure.
Validation: Use subsequent audits to determine whether the error reduces.
The important question is:
Did the action change the result?
Measure Improvement After the Intervention
Suppose incorrect validation accounts for 72 defects in the baseline month.
After corrective action, the next month records 35 defects.
That looks promising.
But management should also check whether:
Production volume changed
Audit volume changed
Sampling methodology changed
Worktype mix changed
Error definitions changed
A reduction in raw count is not automatically proof of improvement.
Where possible, compare a consistent error rate, such as:
Audits with the specified error ÷ Relevant audits × 100
using comparable populations.
That provides a stronger basis for evaluating whether the intervention worked.
Pareto Analysis Should Be Repeated
A Pareto chart is not a one-time exercise.
Quality priorities change.
One error category may decline after coaching.
Another may emerge after a new product launch.
A policy change may introduce a new failure pattern.
Regular Pareto analysis helps management identify whether the concentration of defects is changing.
The process becomes:
Audit
↓
Classify Errors
↓
Analyze Pareto
↓
Investigate Root Cause
↓
Take Corrective Action
↓
Measure Improvement
↓
Review the New Pareto
This is how quality data becomes part of continuous improvement.
Common Pareto Mistakes
A few mistakes can make the analysis misleading.
Treating 80/20 as a rule: The data does not need to produce exactly 80%.
Ignoring severity: Rare critical errors may deserve attention even when their frequency is low.
Using inconsistent categories: Overlapping error definitions distort the ranking.
Ignoring denominators: A high count may reflect a much larger audited population.
Mixing sampling methods without context: Targeted sampling can change the observed error distribution.
Assuming correlation is root cause: Pareto identifies concentration, not causation.
Stopping at the chart: The value comes from investigation and corrective action.
What Should a Quality Dashboard Show?
A useful quality dashboard should allow management to move from the overall result into the underlying detail.
For example:
Overall Quality
↓
Error Categories
↓
Pareto Ranking
↓
Severity
↓
Worktype / Team / Employee
↓
Individual Audit Findings
This allows leaders to answer:
What is driving the result?
Where is it concentrated?
How serious is it?
What should we investigate first?
That is much more actionable than a single headline percentage.
From Quality Reporting to Quality Intelligence
A basic quality report says:
Quality = 93%
A better report says:
Quality = 93%, and three error categories account for 76% of observed defects.
A stronger quality-management process goes further:
Those errors are concentrated in two Worktypes, and the most significant category increased after a procedure change.
Now management has a specific investigation to perform.
The goal is not simply to produce more reports.
It is to make quality data useful for operational decisions.
How Praevexa QualityFlow Can Help
Praevexa QualityFlow is designed to help quality teams move beyond recording audit scores toward structured quality analysis.
QualityFlow supports capabilities such as configurable QA Worktypes, audit checklists, error categories and severity, sampling, Pareto analysis, quality trends and hierarchy-based reporting.
These capabilities help organizations examine quality performance by process and identify where defects are concentrated.
The objective is to move from:
Audit → Score → Report
toward:
Audit → Classify → Analyze → Investigate → Improve
Learn more about Praevexa QualityFlow.
Related Reading
Why Quality Scores Alone Don’t Tell You Where the Process Is Failing
What Should a Quality Management System Actually Do? 12 Capabilities Beyond QA Scoring
QA Sampling: How Much Should You Audit and How Should You Select the Sample?
Critical vs Major vs Minor Errors: How to Design a Better QA Error-Severity Framework
From QA Finding to Improvement: How to Build a Closed-Loop Quality Feedback Process