Split a CSV file
Break a large CSV into files of a set number of rows, into equal parts or one file per value in a column. The heading row goes on every file.
Large files are fine; a few hundred megabytes takes a few seconds.
Your file is split on this device. Nothing is uploaded.
Why split a CSV
Mostly because something downstream has a limit. Excel stops at just over a million rows. Email platforms, CRMs and ad platforms cap the size of an import, often at a few thousand or tens of thousands of rows or at a number of megabytes. Some APIs take batches of a fixed size. Splitting by row count turns one file that is refused into several that are accepted.
Three ways to split
Every N rows gives files of a fixed size, with the last one holding the remainder, which is what import limits call for. Into N equal files divides the rows as evenly as possible, useful for sharing work between people.By column value groups rows by a field, so an export of every order becomes one file per country or per customer.
Putting it back together
The parts keep the original order and the same columns, so they can be recombined exactly with the CSV merge tool. File names are numbered with leading zeros, so they sort in the right order in any folder.
Common questions
Does every file get the heading row?
Yes, when First row is headings is ticked, which it is by default. A CSV part without its headings is almost useless to whatever imports it, since the columns would have no names.
Can a row be cut in half?
No. The file is parsed properly before it is split, so a quoted value containing a line break, common in address and notes columns, is treated as part of its row. Splitting by counting lines, as simple tools do, cuts those rows in two.
How big a file can it split?
Hundreds of megabytes on a typical laptop, limited by the memory your browser allows a tab. Files of that size are exactly the ones Excel refuses to open, since it stops at 1,048,576 rows.
How does splitting by column work?
Choose a column, such as region or customer and every distinct value gets its own file containing only its rows, named after the value. It is the quickest way to send each branch, client or salesperson their own slice of an export.
Guides
The other tools
All of them work the same way: the file is read in your browser and never uploaded.