Column Statistics
Profile every column: count, mean, median, min, max, unique.
About Column Statistics
Generate a statistical profile of each column — non-empty count, distinct values, minimum, maximum, sum, mean, median, standard deviation for numbers, and the most frequent values for text — with a distribution sparkline.
Before analysing a dataset it is worth knowing what is in it — how complete each column is, how varied, and whether the numbers fall where you expect.
A statistical profile of every column at once.
- Processed on your deviceEverything you open or type here is handled by this page and never uploaded anywhere.
- No queue, no limitsNothing waits on a server, so the only limit is your own hardware.
- No account neededNo sign-up and no email address required.
How Column Statistics works in your browser
Each column is scanned and profiled: non-empty count, distinct values, and for numeric columns the minimum, maximum, mean and median. The profile is computed over the whole column rather than a sample, so the numbers describe your actual data.
The useful signals are usually the boring fields. A non-empty count well below the row count means missing data that will skew any average taken over it. A distinct-value count of one means a column carrying no information at all. A distinct count equal to the row count means an identifier rather than a category.
Minimum and maximum are the fastest way to find data-entry errors. A negative age, a date in 1900, or a price three orders of magnitude above the rest are all visible immediately and effectively invisible when scrolling.
When to use it
Sizing up a dataset somebody has just sent you before doing anything with it. Finding the columns with enough missing values to invalidate an analysis. Spotting outliers and impossible values caused by data entry. Documenting the shape of a dataset for someone else who will use it.
Read more about this
Frequently asked questions
What is reported for each column?
Non-empty count, distinct values, minimum, maximum, sum, mean, median and standard deviation for numeric columns, and the most frequent values for text columns.
How does this help before analysis?
It shows the shape of the data at a glance — an unexpected maximum, a column with one distinct value, or far fewer non-empty cells than rows all point at problems worth fixing first.
Are text columns useful here?
Yes. The most frequent values reveal inconsistent categories, such as the same country spelled three different ways.