Split CSV

How it works

A typical case: you have an order export or a customer list with 250,000 rows, but your store's bulk import accepts at most 10,000 rows, your email service takes 50,000 contacts per upload, or Excel freezes while opening the file. Tap or click "Choose a CSV file", pick the file, type the limit into "Data rows per file" and press "Split into files". With the 10,000 setting, customers.csv with 25,000 data rows becomes customers_1.csv (10,000), customers_2.csv (10,000) and customers_3.csv (5,000), all three inside customers_parts.zip. The header row is not counted and, by default, is the first line of every part, so each file can be imported on its own. The split follows records, not lines: an address cell that holds a line break inside quotes is never cut in two. The file is not rewritten, only sliced: commas or semicolons, quotes, values such as "1,250.50" and the encoding (UTF-8 or Windows-1252) stay identical byte for byte, and a UTF-8 BOM at the start of the file is added to every part. If the file already fits in one part, you get it back as it is.

This tool is also known as split csv, csv splitter, split large csv file, split csv into multiple files, split csv by rows.

What is CSV?

CSV is a plain-text table in which each line is a record and cells are separated by a comma or a semicolon. Being plain text, it has no size limit: producing an export with millions of rows is easy, but the program that has to open it is not always as relaxed. The rule that matters for splitting: a cell in double quotes may contain a line break, so a record ends at a line break outside quotes, not at every line break.

What is ZIP?

ZIP is a format that bundles several files into one archive and usually compresses them. This tool puts all parts into a single ZIP, so your browser does not try to download dozens of files one by one and you save just one file. CSV is plain text and compresses well; the archive is often several times smaller than the original. Windows, macOS, Android and iOS open ZIP files without extra software.

What is the difference between splitting a CSV and splitting it into sheets in Excel?

To split data into sheets in Excel, the file must open first; that fails with a file over 1,048,576 rows or one that slows your computer down, and saving still gives you a single .xlsx. Splitting the CSV cuts it into separate files without opening it: each part can be uploaded to another system on its own, sent by email or opened on another computer. The result is still CSV, so the format that import screens ask for does not change.

Why the line count sometimes looks off: line breaks inside quotes

In a CSV, a "line" and a "record" are not always the same thing. An address or notes cell in quotes can contain a line break: "123 Main St" and "Apt 4B, Boston, MA 02134" may sit on two lines yet be a single cell. Simple tools that split by counting lines cut that record in half; the next part starts with a stray quote and every column shifts. This tool tracks quotes and cuts only where a record ends. So when you open a part in Notepad you may see more lines than the number you set, while the record count is exactly what you set.

The quote rule is the same one Excel and other CSV readers follow: a quote opens only at the start of a cell, and "" inside it is one literal quote mark. A quote in the middle of a cell, as in 27" monitor, is data. If the file has a quote that never closes, the result line warns you, because everything after it was counted as one record.

• The header row is not counted toward the part size
• Blank lines are not counted but stay in the file as they were
• CRLF (Windows), LF (Mac/Linux) and old CR line endings are recognized
• Part names sort correctly: with 10 or more parts they are name_01, name_02…

How many rows to choose, and should every part get the header?

Base the row count on the limit of the place you will upload to, and leave a little room: if a panel says 10,000 rows and counts the header as a row, type 9,999. An Excel sheet opens 1,048,576 rows, but on a slow computer files above about 100,000 rows are already sluggish; for reviewing, 50,000 to 100,000 rows is a comfortable part. Some email and SMS services limit file size rather than rows (10 MB, for example); in that case divide the file size by the total in the result line to get the average size of a row.

"Repeat in every file" makes each part independent: each one starts with the column names and can be imported alone. If you will join the parts again with a program later, choose "First file only" so the header does not reappear in the middle of the combined file.

Frequently asked questions

How do I split a large CSV file?

Tap or click "Choose a CSV file" on this page and pick your file. Type how many data rows each part should have and press "Split into files". The parts, named name_1.csv, name_2.csv and so on, download to your Downloads folder in one ZIP file; double-click the ZIP to open it.

Excel will not open a CSV with more than a million rows. What do I do?

Excel shows at most 1,048,576 rows on a sheet; it leaves out the rest and warns that the file was not loaded completely. Split the file with this tool into parts of, say, 500,000 rows; each part is a separate file that opens in Excel in full. Because the header repeats in every part, each file keeps its column names.

Will accented letters or decimals get damaged in the split files?

No. The tool does not rewrite the file; it slices the original bytes, so the encoding, the separator, the quotes and values like 1,250.50 stay exactly the same. If the file starts with the UTF-8 BOM that Excel uses to recognize letters such as é or ñ, that mark is written at the start of every part too.

Is my file uploaded to a server?

No. The file is read and split piece by piece inside your browser; a customer list or order export never leaves this page. With very large files (hundreds of MB), building the ZIP can take a few seconds, because all the work runs on your own device's processor.