Friday, June 10, 2011

Classical decision making techniques: Pareto Analysis


"Men stumble over the truth from time to time, but most pick themselves up and hurry off as if nothing had happened."

Winston Churchill

This technique uses a variation of the Pareto principle i.e. “80% of the advantage comes from 20% of the work”. In many situations changing 20% of the situation can resolve 80% of the problems. If we can identify a small number of changes we can make which will make a large difference then we can have a large impact with a small amount of effort.

This approach is particularly useful in innovation scenarios where a quick short term gain can produce and easy and early win and create buy in to the innovation process.

using the technique

  • Create a list of the factors that may be responsible for causing your desired outcome.

  • When you are reasonably sure that you have most of the significant factors listed create a second column with an approximate percentage estimate of each factor's contribution to the outcome.

  • Starting with the largest percentage and working downwards from a percentage point of view create a running total of the percentages. When you have reached about 80% then stop. You have now identified the most significant factors.


This technique does not always work. If you find that you have many factors all with relatively small significance percentages then most of the factors will need to be included to get to the 80% mark. In this case the Pareto analysis may be of little use to you. In cases where you have only a relatively small number of factors (about 20%) needed to make up the 80% significance, this is where this analysis is the most useful.

You should also be cautious of factors that grow over time, they can start of with a small significance but one which gets larger as time progresses.


If you enter the factors and percentages in a spreadsheet it is very easy to sort the spreadsheet by the percentages column and create another column for cumulative percentages.

making your decision

Once you have established the 20% (or so) of the factors that have the most (80%) significance then you can implement those ignoring the others. This will hopefully give you the vast majority of the advantage of the decision with a small amount of effort.

You of course can run the analysis again after a period of time has elapsed. The Pareto principle should still apply. Of the original factors which are left you should find that 20% of these are responsible for 80% of the remaining issues. Of course because you have already implemented the 20% easiest parts of the decision running the next analysis will not produce as big a difference and will take more work but this is a good approach for a rolling implementation of a decision.

1 comment: