Variations are hidden in averages

We look at averages all the time. This is especially true in business.

  • The average customer call lasts for 5 minutes.
  • The average customer has their query resolved in 24 hours.
  • The average application time is 5 days.

This could be a great performance, especially if the range of performance is very tightly clustered around the mean. But unfortunately, we don’t routinely look at ranges to understand their relationship with the average.

Let’s take a look at what I mean.

This company requires you to apply for one of their products. It’s a lengthy process with lots of variables. The companies target is to achieve a 5-day turnaround for its customers.

And guess what, the average application process time is 5 minutes! Everyone give yourself a bonus.

It tells us that just 50% of customers are actually getting an even better experience, with processing times ranging from 1 to 2 days.

But just under half the customers are getting a much worse experience.

At least a 3rd of customers are not receiving the 5-day target the company is aiming for.

Could they be doing better? Probably.

Without understanding the variability of the application process better, we couldn’t say for sure. But I’d begin by looking at why that 3rd of applications took longer than 5 days and identifying the root cause.