I can't remember where I first heard it, but "averages hide the extremes" can be a really useful guide when thinking about how you manage data. Here's a nice illustration from one of my clients:

As they looked at their event registration they saw an average of 100 registrations per month over the course of six months. But a closer look showed that the first month of registration and the last month of registration was where the bulk of those registrations came in. In fact, if they removed the first and last month, the average number of registrations per month was actually more like 50.

The same also happens frequently with membership joins and renewals, certification, and other types of sales.

What this means is averages can be misleading. Basing a decision (e.g., staffing, technology, business processes) on the average number of transactions may lead to some choices that won't serve you well. So be sure to dig deeper when discussing averages.

And on a related note: Using the extremes (rather than the averages) can also be misleading because "extremes" is just another word for exceptions.