Pareto Analysis

5 MINUTES READ

Also known as The Law of the Vital Few and The 80/20 principle.

Pareto is a principle that helps to focus on the most important matters. It describes a phenomenon that a small number of high values contribute more to the total than a high number of low values. The main idea behind this principle is to identify the “vital few” from the “trivial many” to be able to focus and obtain the maximum benefits. Pareto analysis is one of the seven basic tools of quality and considered one of the key tools in Six Sigma and Total Quality Management.

The Pareto principle states that roughly 80 percent of the results come from 20 percent of the efforts. And in the field of continuous improvement, it states that roughly 80 percent of the problems or effects comes from 20 percent of the causes. The exact percentages may vary in each situation. However, few efforts are usually responsible for most of the results, and few causes are usually responsible for most of the effects.

The Pareto principle has been found to be true in many fields and situations. For example, 20 percent of a company’s clients are responsible for 80 percent of its revenue, 20 percent of the population owns 80 percent of the nation’s wealth, the most active 20 percent of twitter users are responsible for 80 percent of the tweets overall, and most importantly, just a few causes account for most of the effect in a fishbone diagram! Equivalent examples from our personal lives might be that you are using 20 percent of your household tools 80 percent of the time.

The Pareto analysis is often used by decision makers to identify the efforts that are most significant in order to decide which to select first. It is also useful during process improvement projects to focus on the causes and factors that contribute most to a particular problem. It can also be used in project management when prioritizing projects to focus on the significant projects that will bring value to the customer and the business.

The Pareto principle can help you measure the impact of an improvement by comparing the before and after. The new Pareto chart should confirm a reduction in the primary causes.

Pareto Chart

A Pareto chart is a special type of bar chart that plots the frequencies of categorical data. The horizontal axis represents the categorical data while the vertical axis represents the frequencies of the categorical data. The bars are arranged in order of frequency from left to right so that the vital few categories can be clearly addressed on the left.

You may optionally have a cumulative frequency curve above the bars, and a right vertical axis to represents the cumulative percentages. This will simplify interpreting the results and make it easier to see the 80/20 connection. And if you have many categories, you may group those small and infrequent categories into an ‘Other’ category and place it on the last bar on the right.

Focus your efforts on those categories whose vertical bars account for about 80 percent

If the resulted Pareto chart clearly illustrates a Pareto pattern, this suggests that only few causes account for about 80 percent of the problem. This means that there is a Pareto effect, and you can focus your efforts on tackling these few causes. However, if no Pareto pattern is found, we cannot say that some causes are more important than others. Make sure that your Pareto chart contains enough data to be meaningful.

Constructing a Pareto Chart

With your team, define the problem and identify the possible causes (using brainstorming for example).
Collect the categorical data to be analyzed.
Calculate the frequencies of the categorical data.
Draw a vertical line on the left of the chart to place the frequencies.
Draw a horizontal line and place vertical bars above it to indicate the frequencies of the categories.
Sort the categories in order of frequency of occurrence with the largest on the left.
Calculate then draw the cumulative frequency curve and the cumulative percentage line.
Label the chart with a title and any other necessary information such as the date and the source of the data used.
If you observe a Pareto effect, focus your improvement efforts on those few categories whose vertical bars account for most.

Example

A factory team has collected data and prepared the following Pareto chart to address the rising number of customer complaints in a way management can understand.

Note how the chart helps visualizing the data to better understand the focus areas.

Example

A team in a hospital has collected data to study the increased number of call bell usage by patients in order to improve nursing services.

The Pareto effect can be seen here to a degree. The first four factors (40 percent) account for about 70 percent of the effect (as shown by the cumulative frequency curve). Consider combining similar categories when possible. For example, it is possible to combine “toilet assistance” and “back to bed” under one category.

Example

A factory team has collected data to address the rising number of customer complaints. Note that the main categories are too generic.

The Pareto chart has further been analyzed and the main categories where specific problems occur most often have been sub-categorized.

The result suggests that there are three sub-categories that occur most often.

In your analysis, consider using contextual data, metadata and the columns that contain textual data. Databases often contain lots of categorical data about the environment from which the data is taken. This data can be very useful in later analysis when investigating the root cause concepts and ideas.