The word “Pareto” has been effectively turned into a buzzword in the business world today. Managers will often request that employees “Pareto” data for a presentation, and in many cases, employees will confuse a Pareto chart with a standard bar graph or pie chart.
In many ways, a Pareto chart is similar to these other graphical data analysis methods, but with a few key distinctions. These distinctions are key to what makes a Pareto chart such a powerful tool for root cause analysis.
In this article, we will explore what exactly a Pareto chart is, how to create a Pareto chart, and the advantages and disadvantages of using a Pareto chart for root cause analysis.
Key Takeaways
✔️ Pareto charts are a root cause analysis tool that can be used to determine the most impactful actions that you can take to resolve a particular problem.
✔️ The basis of the Pareto chart is the Pareto principle or 80/20 rule. Essentially, this means that 80% of the impact comes from 20% of the work.
✔️ A Pareto chart is different from a bar graph because it is organized in descending order and has a cumulative percent line and 80% cut-off line.
✔️ You can take an iterative approach when using a Pareto chart to dig deeper into the root cause of the problem.
What is a Pareto chart?
A Pareto chart is a type of bar graph where the height of the bars correlates to the frequency or impact of various issues in a business or process. Some examples of this are:
- The frequency of customer complaints regarding a particular issue
- The frequency of a particular defect
- The frequency of a particular equipment failure
- The cost of a particular defect or issue
Then, these bars are arranged in descending order so that the most critical issues are listed on the left-hand side of the chart and the more trivial issues fall on the right-hand side. This sorting method is one of the key attributes that distinguishes a Pareto chart from a simple bar graph.
Another element that is key to the creation of a Pareto chart is a cumulative percent line. This means that the Pareto chart is a dual y-axis chart. The various axes represent the following:
- The x-axis shows categorical data that is typically listed as the name/cause of an issue. For example, one category may be “Dents” if you are creating a product where dents may occur during the production process.
- The left y-axis gives the frequency or impact of the issues. This is typically given as a number. For example, if dents were observed 5 times in the process over a selected time period, the y-axis would correlate to 5 for the bar labeled “dents” on the x-axis.
- The right y-axis gives the cumulative total of data related to each bar in the chart. This is given as a percent. You can see an example of this calculated in the figure below.
The final component of a Pareto chart is the 80% cut-off line. This line is key in visually demonstrating the 80/20 rule or Pareto Principle in your chart.
Any bars with cumulative percent data points that fall under the 80% cut-off line are considered to be the “vital few”. Conversely, any bars with cumulative percent data points that land above the 80% cut-off line are considered to be the “trivial many”.
The “vital few” are contributing to 80% of the problem, therefore, they should take the highest priority when determining areas to improve as they will give you the “biggest bang for your buck”. Since the “trivial many” only represent 20% of the problem, you will not see much improvement if you focus on these areas. This is the idea behind the 80/20 rule which we will dive deeper into below.
What is the 80/20 rule?
The 80/20 rule or Pareto principle was first mentioned by Italian economist Vilfredo Pareto in his 1896 book “Cours d’economie politique”. In this instance, Pareto discovered that 80 percent of wealth was held by only 20 percent of the population.
From this first observation, the Pareto principle was found to be a pattern that applies in basically every situation:
- 80% of errors come from 20% of machines
- 80% of work is completed by 20% of the workforce
- 80% of bad raw materials come from 20% of suppliers
- 80% of the benefits come from 20% of the work
- 80% of complaints come from 20% of customers
- 80% of the problems can be correlated to 20% of the causes
The 80% and 20% figures are not a set rule, but instead more of a guideline. The exact values fluctuate up and down depending on the situation, but the overall outcome remains the same in nearly all situations: very few causes lead to most of the problems.
Therefore, in business, a Pareto chart will help you harness this Pareto principle to discover what few critical causes (the “vital few”) you must address to eliminate the majority of your problems.
When should you use a Pareto chart?
As discussed so far throughout this article, Pareto charts are a powerful tool for determining the “vital few” issues that you need to address in order to achieve the greatest benefit for your business. Therefore, Pareto charts are an extremely beneficial starting point during root cause analysis and project selection.
Pareto charts should be used when:
- There is a recurring failure in a process, and you need to determine the cause.
- You want to select the most impactful issue to resolve first in order to improve profits.
- You want to communicate the most critical issues in a process clearly to other stakeholders.
- A change was recently made to a process, and you want to analyze the improvement after the change to determine the solution’s effectiveness.
So far, this article mainly highlights the usage of Pareto charts in business and manufacturing, but you can use this powerful tool in any industry or even in your personal life. In the next section, we will walk through the steps for creating a Pareto chart along with an example related to personal life to demonstrate how well this tool translates to all areas of life.
How do you create a Pareto chart (with an example)?
STEP #1 IDENTIFY AN ISSUE YOU WANT TO ANALYZE/RESOLVE
The first step towards creating a Pareto chart may seem obvious, but it is critical to defining the scope of your investigation. In this first step, you must identify the particular issue that you want to analyze and resolve.
🔎 EXAMPLE: Let’s say you are consistently late to work. In this case, the issue that you want to analyze and resolve is your tendency to be late for work.
STEP #2 DETERMINE YOUR METRICS
Next, you must determine the appropriate metrics to use. Most commonly this will be one of the following:
- Frequency of occurrence of a particular issue/cause
- Cost
- Time Lost
As a part of this step, you must also determine the time range over which you plan to complete this analysis. Be sure that you are allowing enough time to collect a representative set of data to accurately depict the situation.
🔎 EXAMPLE: If you want to determine the cause of your tendency to be late for work you may use the frequency metric to analyze the frequency with which various events occur that cause you to be late. You may also decide to analyze this situation for four months to get an accurate conclusion of the primary cause(s).
STEP #3 DEFINE YOUR CATEGORIES
Then, you must define the possible causes for the issue that you wish to analyze and resolve. These will be the categories that you will use to tally your data as you collect it. The following are a few examples of how to gather a list of categories:
- Experience with the process
- Process data
- Customer surveys
- Employee feedback
- Customer complaint history
One important note is to include an “Other” category in this list. This will account for any categories that you may not have thought of.
🔎 EXAMPLE: The following are reasons that you may list as to why you have been late for work:
- Overslept
- Distracted by your children
- Traffic
- Bad weather conditions
- Other
STEP #4 COLLECT DATA
After you have defined your metrics and categories, you are ready to begin collecting data! Throughout the time period that you decided to analyze, be sure to tally the cause of the issue each time that it occurs.
🔎 EXAMPLE: You recorded the cause each time that you were late for work over the past four months, and had the following findings:
| CATEGORY | NUMBER OF OCCURRENCES |
| Overslept | 30 |
| Distracted by your children | 5 |
| Traffic | 10 |
| Bad weather conditions | 3 |
| Other | 2 |
STEP #5 GRAPH THE DATA IN A BAR GRAPH FORMAT
Once you have collected data over your selected timeframe, you must then graph this data. This is most easily done in a program such as Excel.
The first step of this is creating a bar chart where the x-axis labels the category name for each bar and the y-axis represents the metric that you selected in Step #2. The bars should be sorted in descending order.
Also, the y-axis should extend from 0 to the total number of occurrences observed, total cost, or total time lost depending on the metric you defined in Step #2. This ensures that you do not overestimate the impact of a certain cause based on the scale of the graph.
🔎 EXAMPLE: When you create a bar graph using the data collected on reasons why you were late to work in the past four months, the x-axis should contain the labels for each of the categories defined in Step #3 and the y-axis should represent the frequency of occurrence of each of those events. Make sure that the bar graph is sorted in descending order as shown in the figure below.
Also, the y-axis has a scale from 0 to 50 since you were late to work a total of 50 times over the four-month period.
STEP #6 ADD A CUMULATIVE PERCENT LINE
Next, you will need to add a cumulative percent line to the chart. To do this, follow these steps:
- Calculate the percentage of each category. This is the subtotal of each category divided by the total of all the categories.
- Calculate the cumulative percentage. For the first category on the graph this will be equivalent to its percentage calculated in the previous step. Then, the cumulative percent for the second category will be the percentage of the first category added to the percentage of the second category. Next, the cumulative weight of the third category will be the percentage of the first category plus the percentage of the second category plus the percentage of the third category. And so on.
- Add a secondary y-axis for the cumulative percent to the Pareto chart. The Pareto chart should currently have a left y-axis that represents the metric selected in Step #2. You must now add a secondary y-axis on the right side of the Pareto chart which will represent the cumulative percent. This axis should be labeled from 0% to 100%.
- Add the cumulative percent line to the Pareto chart. Above each bar in the chart, add a dot representing the cumulative percentage for that category as calculated earlier. Once this is complete, draw a line to connect these dots. This is the cumulative percent line. The final dot should reach 100% on the right y-axis.
🔎 EXAMPLE: The Pareto chart showing the reasons why you were late to work is given below with a cumulative percent line.
STEP #7 ADD AN 80% CUT-OFF LINE
The final step to complete your Pareto chart is to add an 80% cut-off line. This line is a visual representation of the 80/20 rule or Pareto principle. Any categories with cumulative percent data points that fall under the 80% cut-off line are called the “vital few”. The remaining categories are the “trivial many“. The “vital few” categories are the ones you should focus your efforts on in order to see the largest impact.
🔎 EXAMPLE: The finished Pareto chart showing the reasons why you were late to work is given below.
In this case, the “vital few” categories are oversleeping and traffic. All the other categories fall into the “trivial many”.
STEP #8 USE THE PARETO CHART TO TAKE ACTION
Now that you have a completed Pareto chart, the top causes for your issue should be evident. Therefore, you can now focus your attention on mitigating the most impactful causes of your problem instead of getting bogged down in the “trivial many”.
In some cases, you may find that the top cause that you have identified is only the beginning of identifying the true root cause of your problem. If this is the case, you can now use the Pareto method again to dig deeper into the causes of the top issue identified in this first Pareto chart or you can use other root cause analysis tools to help pinpoint a deeper cause for your problem.
🔎 EXAMPLE: Let’s take a look at this in regard to our example about reasons why you were late to work. In this case, the top cause identified in the Pareto chart is that you overslept. This tells you that the main problem here is your habit of oversleeping, but you may now be wondering, what is the main cause of your habit of oversleeping? To dig deeper into this, you can complete another Pareto analysis where you tally the reason each time you oversleep. At the end of this second Pareto analysis, you may find that the reason for your oversleeping, and hence the reason that you are late for work, is that you frequently forget to set an alarm. This results in a root cause that you can take action on by determining a method for ensuring that you always have your alarm set.
Advantages of the Pareto chart
Overall, the advantages of using a Pareto chart are that it:
- Is a data-driven approach to root cause analysis.
- Helps you accurately identify and prioritize actions based on their overall impact.
- Allows you to more effectively allocate your time and resources to high priority, high impact actions.
- Improves your ability to make decisions on key areas of focus.
- Drives continuous improvement in your organization.
- Improves productivity and profitability for your organization.
Disadvantages of the Pareto chart
Although Pareto charts are an extremely powerful tool, there is one major watch-out that you must pay attention to when using a Pareto chart to ensure you accurately interpret the data.
This watch-out is that Pareto charts do not take the cost of solution into account.
When creating a Pareto chart, the highest priority item is determined based solely on which action will have the greatest impact. In some cases, this action may be very expensive, and therefore, the cost to complete the action could be more expensive than the return that you will see once the solution is implemented. Therefore, it is important to complete a Cost Benefit Analysis to ensure that the action you are taking is worthwhile.
One way you can avoid this issue is by using a weighted Pareto chart. In a weighted Pareto chart, each category is assigned a weight based on the cost of the fix. This weight is then multiplied by the occurrence to determine a weighted score for each category, and the Pareto chart is created using this weighted score.
Tips for successfully using a Pareto chart
Now that you know how to create and interpret a Pareto chart, here are a few tips and reminders to ensure that you do so successfully:
- Make sure your categories are not too specific. If your categories are very specific, you may find that each category only has a few occurrences at the end of your analysis and no categories will stand out as the most impactful cause. If you do find that you have a lot of different categories, try putting them into related groups. For example, if you are looking at the number of occurrences of defects on a car, and you have categories for “scratches on the hood”, “scratches on the door”, and “scratches on the roof”, you should consider combining these into a more general “scratches” category.
- Verify that your dataset is statistically significant. Most statisticians consider a dataset to be statistically significant if it has at least 30 data points. Make sure that you are conducting your study over a long enough time period in order to gather a representative, statistically significant dataset.
- Include an “Other” category. This can be used to account for any reasons you may not have thought of before starting your analysis and/or to pool several of the “trivial many” categories together to make your final Pareto chart cleaner.
- Use a weighted Pareto chart for more clarity. Weighted Pareto charts can be weighted either based on the criticality of a specific category or the cost of repair. This will allow you to accurately prioritize actions based on how critical the issue/repair is or how expensive the repair is.
Conclusion
Pareto charts have become common lingo in business today, but many people do not fully understand what they are. Therefore, the power and simplicity of this tool is frequently missing in many organizations. If you have a recurring problem and want to determine the most impactful actions that you can take to resolve it, you should definitely be using a Pareto chart!
Did you have any misconceptions about Pareto charts? I’d love to hear your thoughts in the comments below!
