Free practical guide
Mean vs median: which average should you use?
Mean and median both describe a typical value, but they answer slightly different questions. The mean uses every number. The median identifies the middle of an ordered list.
One dataset, two different answers
Imagine five fictional order values, expressed in thousands of euros: 4, 5, 6, 7 and 28. The first four orders are similar; the fifth is much larger.
| Order | Value |
|---|---|
| 1 | €4,000 |
| 2 | €5,000 |
| 3 | €6,000 |
| 4 | €7,000 |
| 5 | €28,000 |
Calculate the mean
Add the values and divide by how many there are. The sum is €50,000 across five orders, so the mean order value is €10,000.
Mean = (€4,000 + €5,000 + €6,000 + €7,000 + €28,000) ÷ 5 = €10,000.
Every order affects the mean. The €28,000 order pulls it upwards. Removing that single order would leave a mean of €5,500 across the remaining four.
Calculate the median
Put the values in order. With five values, the third value is in the middle. The median is €6,000.
Ordered values: €4,000 · €5,000 · €6,000 · €7,000 · €28,000.
If the dataset had an even number of values, the median would be the mean of the two middle values.
Which number is more useful?
Choose the measure that fits the decision. The mean is useful when you need a total-per-item perspective, for example to reconcile an average with a total. The median may better represent the middle order when a few unusually large values distort the mean.
Neither statistic describes the whole distribution. For a business decision, show the sample size, time period and a range or distribution alongside the average.
Related concepts and next step
Read Insyt's mean and median definitions. Then use Data Literacy for Business Decisions to check whether a reported average actually supports the conclusion being drawn.
Keep learning with Insyt
Follow practical lessons and work with realistic fictional company data. Everything is free to explore, with no account needed.
Explore free Paths