Median

The middle value when data is sorted from lowest to highest—exactly half the values fall above and half below. Unlike the mean (average), the median is resistant to extreme values, making it the better measure of 'typical' in skewed distributions. Median household income, not mean, tells you what a typical family earns—because a few billionaires pull the mean far above most people's reality. Whenever a distribution has a long tail, the median and mean diverge, and the median is almost always more useful.

When to use it

When summarizing data that has outliers or long tails; when the mean seems unrepresentative of typical values; when comparing distributions with different shapes; when making decisions about 'typical' outcomes (what most customers, employees, or projects actually experience).

How it can help

Default to reporting medians instead of means for any distribution that might be skewed—income, company sizes, response times, customer lifetime values. When someone presents an average, ask: what's the median? If they differ significantly, the distribution is skewed and the average is misleading. In business: 'average customer lifetime value' is often inflated by a few whale customers; the median tells you what most customers are actually worth.