How to Lie with Averages


I hate averages.

I just hate them.

There... I said it.

Here’s why.

Averages often lie.

Let me illustrate my point with an example.

Ten people live on the street.

One person on the street earns $10 million a year.

The other nine people on the street earn $0.

On average (did I mention I hate that word?), every person on the street is a millionaire.

On one hand, it’s true that the arithmetic mean is $1 million.

On the other hand, the conclusion is completely misleading... and therefore a kind of “lie.”

One of the very first things I teach in the case interview is to segment or dis-aggregate a number that’s a total or an average.

If I say the total combined income of everyone on the street is $10 million, you should immediately disaggregate the number.

When you do, you realize there are nine people who earn $0 a year, and one person who earns $10 million a year.

(In consulting terms, there are two distinct segments — the broke and the deca-millionaire.)

The disaggregated number provides insight.

Not only does a total or average not provide useful insight, it often leads to incorrect conclusions that seem like insights.

Let me give you another example.

Let's say you run a sales team where the team members on average (ugh...) achieve 110% of their sales goal each month.

They achieved over the goal.

Conclusion: This is a good thing... right?

In this case, half of the sales team achieved 160% of their goal (which is, of course, excellent), while the other half of the sales team achieved only 50% of their goal (which is terrible).

If you only look at the average, you conclude the team performed well. Your next step in managing this team might be to leave well enough alone.

This is the incorrect conclusion.

When you disaggregate or segment the average, you discover that half the team performed spectacularly, while the other half performed terribly.

This would lead the curious sales manager to ask any number of questions:

1) What are the 160% achievers doing that the 50% achievers are not? Can whatever’s working well be replicated or expanded in some way?

2) Can the 50% achievers improve their performance? If so, how?

3) If the 50% achievers can’t improve their performance, should they remain on the team?

When it comes to managing operations in industry, better questions ultimately translate into better decisions.

This is one of many, many ways that you can out-execute a competing company.

It is this kind of thinking that separates the good CEOs from the poor ones. Good CEOs think this way. Good CEOs hire direct reports that think this way.

For more articles like this one on how to be an effective CEO, just complete the form below to be updated on future articles and resources on CEO-Level Skills.

0 comments… add one

Leave a Comment