CSV vs TSV: What's the Difference?

If you work with data, you have almost certainly run into both CSV and TSV files. They look nearly identical when you open them, they store the same kind of information, and they are used for many of the same tasks. So what actually separates them, and when should you reach for one over the other? The short answer comes down to a single character.

What is a CSV file?

CSV stands for comma-separated values. It is a plain-text format that stores a table, one row per line, with a comma between each field. A small CSV might look like this:

name,city,role
Ada,London,Engineer
Grace,New York,Scientist

The first line is usually a header naming the columns, and every line after it is a row of data. Because it is just text, a CSV can be opened by almost anything — spreadsheets, databases, programming languages, and text editors all understand it. That universality is the reason CSV became the default way to move tabular data between systems.

The catch is the comma itself. Plenty of real data contains commas — addresses, full names like "Doe, Jane," prices, or free-text notes. When a value contains the delimiter, the format has to escape it, which is why CSV files wrap such values in quotes:

name,note
Ada,"Born in London, England"

That quoting works, but it adds a layer of rules. Stray quotes, unescaped commas, and inconsistent handling between tools are a common source of broken imports.

What is a TSV file?

TSV stands for tab-separated values. It is the same idea as CSV, but it uses a tab character between fields instead of a comma. The data above as a TSV would be:

name	city	role
Ada	London	Engineer
Grace	New York	Scientist

The logic, the header row, and the one-line-per-row structure are all identical. The only thing that changed is the separator. TSV files usually carry a .tsv extension, and occasionally .tab.

The reason people choose tabs is simple: real-world data almost never contains a tab character. Names, addresses, and notes are full of commas, but you rarely type a tab inside a value. That makes the tab a cleaner boundary marker — when a delimiter never shows up in your data, you almost never need quoting to protect it.

The practical difference

Strip away the jargon and the distinction is this: CSV separates fields with a comma, TSV separates them with a tab. Everything else follows from that one choice. The comma is a common character in ordinary text, so CSV leans on quoting rules to keep data intact. The tab is rare in ordinary text, so TSV can usually skip the quoting entirely.

Here is how the two stack up side by side.

	CSV	TSV
Delimiter	Comma (,)	Tab character
Full name	Comma-separated values	Tab-separated values
Common extension	.csv	.tsv, .tab
Quoting needed?	Often — commas appear in real data	Rarely — tabs almost never appear in data
Tool support	Nearly universal	Wide, but slightly less common
Best when	Sharing broadly; maximum compatibility	Data contains commas; you want fewer quoting headaches
Main downside	Quoting rules can break imports	Tabs are invisible and easy to mangle in editors

Neither column is the winner. They are tuned for slightly different situations, and a good chunk of data work involves moving between them.

When to use CSV

Reach for CSV when compatibility matters most. If you are exporting a file for someone else to use, uploading to a service, or feeding data into a tool you do not control, CSV is the safer bet because virtually everything accepts it. It is the lingua franca of tabular data — when in doubt, a CSV will almost always be understood on the other end. The cost is that you take on the quoting rules, which can trip up data full of commas if a tool handles escaping poorly.

When to use TSV

Reach for TSV when your data is full of commas and you would rather not deal with quoting. Datasets heavy on addresses, descriptions, monetary values, or free text are exactly where the tab delimiter shines, because a tab cleanly ends each field without any escaping. TSV is also popular in bioinformatics, log processing, and other technical pipelines where data is passed between scripts. The one thing to watch: tabs are invisible on screen, so a careless editor can silently turn a tab into spaces and quietly corrupt the structure. Using a viewer that shows the raw delimited data avoids that trap. See how to open TSV files for a safe way to inspect them.

If you want a deeper look at how separators work across formats — including semicolons and pipes — CSV delimiters explained covers the full picture.

How to convert between CSV and TSV

Because the two formats differ only in the delimiter, converting between them is straightforward: you open the file, then re-save it with a different separator. You are not transforming the data, just swapping the character that sits between fields.

CEESVEE makes this a two-step job. Open your file — it auto-detects the delimiter and encoding, so a comma file and a tab file both load correctly without any setup. Then use Save As and choose the export delimiter you want: a comma to write CSV, or a tab to write TSV. The same dialog lets you set the encoding, quoting style, line endings (LF or CRLF), and BOM, so the file you produce is exactly the format the next tool expects. For a step-by-step walkthrough, see converting CSV to TSV.

Because CEESVEE keeps the data in a fast Rust core with a virtualized grid, this works just as smoothly on a 1,000,000-row, 100 MB+ file as it does on a tiny one — there is no practical row limit, and nothing ever leaves your machine.

The bottom line

CSV and TSV are two takes on the same idea: a plain-text table, one row per line, fields split by a delimiter. CSV uses a comma and wins on universal compatibility. TSV uses a tab and wins when your data is full of commas and you want to skip quoting headaches. Knowing which to use — and being able to convert freely between them — means the format never gets in the way of the work.

Download CEESVEE to open, edit, and convert CSV and TSV files for free. It is open source (MIT), 100% local, and built to handle delimited files of any size faithfully. You can also browse the source on GitHub.