Descriptive Statistics for Every Column of a CSV or Excel File
Upload a table and get mean, median, standard deviation, missing values and distribution of every column — free, no signup.
| Name | Age | City |
|---|---|---|
| Ana | 34 | Rio |
| Bruno | empty | Recife |
| Carla | 29 | Rio |
| Column | Type | Empty | Summary |
|---|---|---|---|
| Name | text | 0 | 3 unique |
| Age | number | 1 | mean 31.5 |
| City | text | 0 | Rio (2) |
One line per column: type, missing values and a quick summary.
How to Profile Your Table
A complete guide to configuring your data pipeline.
Step 1: Upload Your Table
Drop a CSV, TSV or Excel file into the upload box. Keep "File has a header" checked if the first row contains column names. Numbers written with a decimal comma (1.234,56) or with currency symbols (R$ 10,50, $1,200) are recognized automatically.
Step 2: Open Table Profile
The Table Profile tool opens on the right. Click Show profile. Nothing in your table is changed — the profile is a read-only report.
Step 3: Read the Report
Every column gets one line with its detected type (number or text), how many cells are filled or empty, and how many distinct values it has. Numeric columns also show the minimum, mean, median, maximum and standard deviation plus a small histogram of the distribution. Text columns list their most frequent values. Use Copy result to paste the summary into a report or e-mail.
Technical Specifications & FAQ
Profiling a table is the first step of any analysis: it shows at a glance which columns have missing values, which ones were exported as text when they should be numbers, and whether a variable has outliers or a skewed distribution. Doing this column by column in a spreadsheet means writing AVERAGE, MEDIAN, STDEV and COUNTBLANK formulas over and over.
flowingTable computes the profile with pandas on the server. A column is treated as numeric when at least 95% of its filled cells can be read as numbers; the standard deviation is the sample standard deviation (n − 1), the same value returned by Excel's STDEV.S and R's sd(). The report also counts fully duplicated rows, so you know whether a de-duplication step is needed before analysis.
Frequently Asked Questions
Which statistics does the table profile calculate?
For every column: detected type, filled cells, empty cells and number of unique values. For numeric columns it adds minimum, mean, median, maximum, sample standard deviation and a histogram. For text columns it lists the five most frequent values with their counts. It also reports the number of fully duplicated rows in the table.
Does the profile change my data?
No. The profile is a read-only report shown in a pop-up window. Your table and your flow of steps stay exactly as they were, so you can profile the table before and after cleaning it to check the effect of each step.
Why is a column with numbers shown as text?
A column is classified as numeric when at least 95% of its filled cells can be parsed as numbers. If more cells than that contain notes such as 'n/a', '<0.01' or units like '12 kg', the column is reported as text. Use the Find & Replace or Handle Missing Values tools to clean those cells, then run the profile again.
Is the standard deviation the sample or the population value?
It is the sample standard deviation, which divides by n − 1. This is the value returned by Excel's STDEV.S function, R's sd() and pandas' std(), and is the appropriate choice when your rows are a sample of a larger population.