How to Run a Pareto Analysis on Manufacturing Defects
To perform a Pareto analysis on manufacturing defects, collect defect counts broken down by reason code over at least two weeks, sort them from highest to lowest frequency, calculate the cumulative percentage for each, and identify the defect types that together account for 80% of total scrap — these are the "vital few" that deserve your corrective action focus.
Why the 80/20 Rule Applies to Manufacturing Scrap
The Pareto principle — named after Italian economist Vilfredo Pareto, later popularized in manufacturing by Joseph Juran — states that a small number of causes tend to produce the majority of effects. In manufacturing quality, this pattern holds consistently: roughly 80% of scrap events come from 20% or fewer of the possible defect types.
This is not a precise mathematical law. In some operations it is 70/30; in others 90/10. The point is the asymmetry: a few defect types cause most of the damage. Spreading corrective action effort equally across all defect types is therefore wasteful — the return is far higher when effort is concentrated on the vital few.
Pareto analysis is the tool that reveals which defect types are in the vital few for your specific operation. Without it, quality managers often chase the most recent or most visible problem rather than the one with the greatest cumulative impact.
Step-by-Step: Running a Pareto Analysis
Collect defect count data by reason code
The foundation of a useful Pareto analysis is defect data tagged by specific reason code — not a single "scrap" tally. Each scrapped part or batch must be classified by cause: dimensional rejection, surface scratch, weld crack, contamination, and so on. Collect this data for a minimum of two weeks; 30 days is ideal to smooth out daily variation and capture enough events for statistical reliability. If you currently log scrap without reason codes, establishing a standardized list of five to twelve reason codes relevant to your processes is the necessary first step. Pareto Base allows you to define custom defect reason codes and tag every logged event, so your data is Pareto-ready from day one.
Sort by frequency, highest to lowest
Once you have totals for each defect reason code across your collection period, rank them from highest count to lowest. This ordering is critical — the Pareto chart displays defect types left to right from most frequent to least frequent, creating the characteristic descending staircase shape. If you are tracking scrap cost rather than unit count, sort by total cost of scrap by reason code instead; this often changes the ranking and surfaces different priorities, especially when expensive scrapped parts are a small proportion of events but a large proportion of dollar losses.
Calculate the cumulative percentage
For each defect type in rank order, calculate its percentage of the total defect count (or total cost), then add a running cumulative percentage column. The first row's cumulative percentage equals its individual percentage. The second row's cumulative is the sum of the first two individual percentages. Continue down the ranked list until you reach 100%. This cumulative column is what you plot as the line on the Pareto chart — it rises steeply at first and flattens as you move to the less frequent defect types.
Identify the vital few
Draw a horizontal reference line at the 80% mark on the cumulative percentage axis. The defect types whose cumulative percentage falls at or below this line are the vital few — the small number of causes responsible for the majority of your scrap. Everything to the right of the 80% threshold is the "trivial many": addressing them is worth doing eventually, but the return per unit of effort is much lower than fixing the vital few. Your corrective action resources should focus almost entirely on the vital few until significant progress is made.
Launch a focused corrective action campaign
Select the single top contributor from your vital few and launch a structured scrap reduction campaign targeting that defect type. Assign one owner, establish the baseline scrap rate for that defect over the prior four weeks, set a specific reduction target (20–30% is realistic in 8–12 weeks), and run weekly review meetings. Pareto Base Campaign Management automates baseline tracking and progress visualization so the campaign owner always knows whether the trend is moving in the right direction. See the full playbook at Scrap Reduction Campaign Guide.
Worked Example: Stamping Line Defect Data
Below is a worked example from a stamping line that logged 162 scrap events over 30 days across four defect reason codes. This is the raw data before any analysis — the kind of data Pareto Base collects automatically as operators log scrap events.
| Defect Reason | Count | % of Total | Cumulative % |
|---|---|---|---|
| Dimensional rejectionvital few | 82 | 50.6% | 50.6% |
| Surface scratchvital few | 45 | 27.8% | 78.4% |
| Weld crack | 23 | 14.2% | 92.6% |
| Contamination | 12 | 7.4% | 100.0% |
| Total | 162 | 100% | — |
How to read this Pareto table
Dimensional rejection alone accounts for 50.6% of all scrap events. Adding surface scratch brings the cumulative total to 78.4% — just under the 80% threshold. These two defect types together represent the vital few for this stamping line. Weld crack and contamination, while worth monitoring, together account for only 21.6% of events.
The practical implication: if the quality team allocates 100% of corrective action effort to dimensional rejection and surface scratch, they are addressing 78% of their scrap problem. A 25% reduction in dimensional rejections alone would cut total scrap events by approximately 12.5% — a meaningful improvement from a single focused campaign.
On a Pareto chart, this data would appear as four bars descending left to right, with a cumulative percentage line that rises steeply over the first two bars and then flattens. The 80% horizontal reference line intersects the cumulative line between the second and third bars — visually marking the boundary between vital few and trivial many.
Common Pareto Analysis Mistakes to Avoid
Logging scrap without reason codes
A single "scrap" category produces a Pareto chart with one bar — useless for prioritization. Every scrap event must be tagged with a specific defect reason to make Pareto analysis possible.
Using too short a collection period
One or two days of data can be dominated by a single unusual event. Collect at least two weeks of data, ideally 30 days, to get a stable picture of your defect distribution.
Analyzing by unit count when cost varies widely
If your most expensive scrapped part appears infrequently but costs 10× more than the most common defect, a unit-count Pareto will understate its importance. Run a cost-weighted Pareto in parallel.
Targeting multiple vital few defects simultaneously
Splitting corrective action effort across three or four campaigns simultaneously dilutes focus and reduces the probability of success. Pick the single top contributor and drive it down before moving to the next.
Pareto Base builds your Pareto chart automatically
As your team logs scrap events with reason codes, Pareto Base generates and updates your Pareto chart in real time — no spreadsheet rebuilding, no manual calculations.