Liar, Liar… “Average” Liar

The “Average” or “Arithmetic Mean” is one of the most common statistical formulas applied in business performance analysis.

Almost everyone knows how to calculate it… it is just the sum of a list of values divided by the number of occurrences.

However, for some business questions the “Average” can deliver deceiving answers and lead to wrong decisions.

Example: A company has 2 branches and they are facing some customer satisfaction issues because of the queue time that their customers endure.

After a customer survey where almost all mentioned that they accept to wait up to 5 minutes, but if they had to wait in queue for more than 5 minutes, they would be extremely unsatisfied. So, the company decided to evaluate their branches and deliver regular incentives to the best performer.

Branch A had 10 customers and the queue time of each one was:

4 min, 3 min, 5 min, 4 min, 4 min, 20 min, 7 min, 4 min, 4 min and 5 min.

Average queue time = 6 min



Branch B had also 10 customers and queue time of each one was:

1 min, 7 min, 6 min, 4 min, 7 min, 2 min, 8 min, 1 min, 7 min and 7 min.

Average queue time = 5 min

So, the Branch B had the best performance… or maybe not.

Remember that the customer survey told us that if the customers had to wait more than 5 minutes they would be extremely unsatisfied.

From the count of impacted customers perspective:

Branch A: 80% of customers satisfied vs. 20% of customers extremely unsatisfied.

Branch B: 40% of customers satisfied vs. 60% of customers extremely unsatisfied.

So, the Branch A was able to deliver the expected service level to 80% of the customers and the Branch B only delivered it to 40%, therefore impacting negatively much more customers than Branch A.

This is a just a simplified example that shows how the “Average” can “lie” when measuring business performance and that is even more risky in business scenarios where outliers can occur as this one.

The “Average” is very useful in business analysis, but be careful when interpreting it, because it also can “lie” to you.

Always keep in mind that sometimes there are other formulas to measure that are more suitable to answer your business question.