PROBLEM SOLVING - 31.03.2010

80/20 rule

Cause and effect

Many of your daily work problems don’t have a single root cause, while each problem may have several quite separate contributing factors. Financial controllers can use Pareto analysis to analyse this cause and effect.

Pareto analysis is the method of looking at all the root causes of a problem and trying to determine which ones have the greatest frequency. The idea behind the analysis is that an entire collection of potential causes can be broken down into those that seldom happen and those that happen on a more frequent basis.

The 80/20 rule. Roughly 80% of the problems stem from around 20% of the possible causes. It’s important to note that the actual percentages (say, 94% to 6%) may differ a bit from 80% and 20%, but the spirit of the principle still remains true. That is, if you can identify and focus on the most frequent causes, the majority of your problems will be handled.

Identifying the real cause

Step 1. Determine the problem for which you want to investigate causes, and the scope of the investigation. For example, the problem: “We want to determine what are the top causes of project delays of one month or more.” Then for scope: “We are going to analyse all projects that completed in the last year.”

Step 2. Decide what categories it makes sense to analyse. For instance, if you’re investigating the source of major project delays, the categories might be “lack of resources”, “increased scope”, “technical issues” etc.

Step 3. Assemble data in each category for the “frequency of occurrence”. In our project delays example, you’d need to get data by talking to project managers or reviewing project files. For each type of delay, you could count the number of projects for which that category was the biggest source of schedule delay. Or you could define your count as, e.g. the number of projects on which this category caused a delay of at least one week.

Step 4. Input the data into a spreadsheet. Include columns showing the Percentage of Total and Cumulative Percentage (see The next step). Use Data/Sort to sort the data in descending order, i.e. starting with the most common reason down to the most rare.

Step 5. Then look for a pattern indicating a Pareto effect at work: a few categories contributing an overwhelming majority to the total frequencies you charted. Ideally, you will see a clear break point at or around 80% on your Cumulative Percentage line. In some cases you may see a more gradual curve - not exactly a Pareto effect, but still a good indication of the key issues. Or you may find a fairly level curve, with each category contributing almost equally to the total.

Step 6. The first cause to tackle is the one that has the highest score. This one will give you the biggest benefit if you solve it. The options with the lowest scores will probably not even be worth bothering with.

For a Pareto analysis example, visit http://financialcontroller.indicator.co.uk (FC 02.07.09).