CSV Column Extractor
Extract just the columns you need from a CSV file - free, no upload.
What this tool does
Paste your CSV, uncheck the columns you don't want, and the result updates instantly with only the columns you kept.
Reads the header row from your CSV and shows a checkbox for every column it finds.
Rebuilds the CSV using only the checked columns, keeping their original order.
Works without a header row too, labelling columns generically by position.
Handles rows with uneven column counts by padding missing cells with an empty value, rather than throwing an error.
Reports how many columns were kept out of the total found, alongside the row count.
How to use it
- Paste your CSV data.
- Uncheck any columns you don't want to keep.
- Copy or download the result.
Why use this extractor
Wide exports hold many columns you do not need. Picking the columns to keep gives a smaller file for sharing or import.
Why we built it
We built the CSV column extractor to build the checkbox list from your actual header row each time, not a fixed set. Remove repeated rows with the CSV duplicate remover and order the rest with the CSV sorter.
Handy alongside this: the CSV Splitter and the Merge CSV Files.
Who this tool is for
- Analysts keeping only the columns they need
- Marketers trimming exports
- Developers preparing test data
When to use it
- Removing sensitive or irrelevant columns.
- Preparing a smaller file for a tool.
- Selecting fields for analysis.
- Reducing size.
Why the checkboxes are built from your actual header row
A fixed list of checkboxes would only work for one specific CSV shape.
Every CSV file has different columns, so a useful column-picker has to read the columns that are actually in the data you pasted rather than presenting a fixed set. This tool re-reads the header row every time your input changes and rebuilds the checkbox list to match, so it works the same way on a two-column export as it does on a twenty-column one.
Why CSV doesn't have one single, enforced format
RFC 4180 documents a common convention, but it isn't a binding standard, so "the columns" can mean different things for different files.
A CSV file's delimiter, quoting style, line endings and character encoding are all conventions rather than requirements - most tools default to a comma delimiter and double-quoted fields with doubled internal quotes (the RFC 4180 approach), but plenty of real exports use semicolons, tabs, or a different escaping rule instead. That matters for a column extractor specifically, because "column 3" only refers to a stable, meaningful piece of data if the file was split into fields the same way a spreadsheet program would split it. This tool parses comma-separated fields with a quote-aware reader, so a field like "Springfield, IL" wrapped in quotes stays one column instead of splitting into two - but a file that actually uses semicolons or tabs as its separator needs converting to comma-separated form first, since otherwise the whole row is read as a single column.
What you give it, and what you get back
| Item | Details |
|---|---|
| Input | A CSV with or without a header row. |
| Selection | Tick the columns to keep. |
| Output | A CSV of the chosen columns in original order. |
Issues you may run into
| What you see | Why it happens | What to do |
|---|---|---|
| Order is not as ticked | Columns keep original order. | Reorder in a spreadsheet. |
| Header names are generic | There was no header row. | Turn on the header option. |
| An error message appears | The input is not valid for this format. | Fix the problem the message points to. |
Assumptions and limitations
- Only whole columns can be kept or dropped - there's no per-cell filtering.
- Column order in the output always matches the original file's left-to-right order, even if you check boxes in a different order.
Frequently asked questions
How do I extract specific columns from a CSV file for free?
Paste your CSV and uncheck the columns you don't want - the result updates instantly.
Does this work without a header row?
Yes - uncheck "First row is column headers" and columns are labelled Column 1, Column 2, and so on.
Can I reorder the columns in the output?
No - kept columns keep their original left-to-right order from the source file.
Does this handle quoted CSV fields with commas inside them?
Yes - it uses a real RFC-4180-aware CSV parser.
Is my data uploaded anywhere?
No. Everything happens in your browser.
What happens if some rows have more or fewer columns than the header?
A missing cell is filled in as empty in the output, and the checkbox list is still built from the widest row found, so nothing shifts out of place.
Does it work with semicolon- or tab-delimited files?
Not directly - this tool splits on commas, so a semicolon- or tab-delimited export needs its delimiters converted to commas first, or the whole row will read as a single column.