CSV Column Stats Table

Get a side-by-side comparison table of every CSV column free online. See data type, fill rate, and uniqueness per column at a glance before you start any analysis.

Paste CSV data

How to Use This Tool

Paste your CSVAdd your full dataset including the header row

Click Analyze ColumnsEvery column is scanned for type, fill rate, and uniqueness

Scan the comparison tableSpot weak or problematic columns at a glance

CSV Column Stats Table is a free online tool that works entirely in your browser — no signup, no data uploaded, no limits.

One Table to Understand Your Whole Dataset

Before running any real analysis, you need a sense of what you are working with. Which columns are mostly empty. Which look like identifiers. Which are categorical fields with just a handful of repeated values. Checking this column by column in a spreadsheet means clicking through each one individually. This tool generates the entire profile in a single pass.

Each row in the output table represents one column from your source CSV: its detected data type, what percentage of rows actually have a value in it, and how many distinct values it contains. Reading down this table tells you more about your dataset's shape in ten seconds than scrolling through the raw data for ten minutes.

Spotting Problem Columns Early

A column with 30% fill rate is a red flag before you build a report that depends on it — most of your output rows would be missing that information. A column where the unique count equals the row count is functioning as an identifier, even if nobody explicitly labeled it that way. Catching these patterns before analysis saves you from drawing conclusions based on incomplete data.

Fill Rate vs Uniqueness — Reading Them Together

High fill rate with low uniqueness usually means a clean categorical field — gender, status, region. Low fill rate with high uniqueness among the values present often means optional free-text fields like a notes column. Recognizing these patterns helps you decide quickly which columns deserve attention and which can be set aside.

Frequently Asked Questions

Most stats tools give you numbers for one column at a time. This generates a single comparison table showing every column side by side — data type, fill rate, and unique count — so you can scan the whole dataset's shape in one view.

Fill rate is the percentage of rows in a column that have a non-empty value. A column at 60% fill rate means 40% of rows are blank for that field, which is useful to know before building any report that depends on that column.

A column where every value is unique is likely an identifier (order ID, email). A column with very few unique values relative to row count, like a status field with only three possible values, is likely categorical. This shapes how you should analyze or visualize it.

Yes, the table scrolls horizontally for wide datasets. Each column gets its own row in the output table regardless of how many there are.

Indirectly. A column with very low fill rate and low cardinality often carries little useful information and is a candidate for review, though the final decision on whether to keep it depends on your specific use case.