median

Median

The middle value in a sorted dataset — half the values fall above it, half below. Equivalent to the 50th percentile.

When you'd see it: Used instead of the mean wherever data is skewed or has outliers: median household income, median time-to-hire, median deal size, p50 latency. In engineering contexts, percentiles do the same job (the median is the 50th percentile).

Why it matters: The median is robust to outliers in a way the mean isn't. When a few extreme values distort the average, the median describes the typical case more accurately. In right-skewed distributions — most revenue data, most engagement data — the median is almost always the more honest central tendency.

Common mistakes: Treating mean and median as interchangeable without checking the data distribution. If they're far apart, the distribution is skewed — and that gap itself is telling you something important about the data.

Related mean, percentile, normal distribution, outlier, descriptive statistics