CSV viewer — open CSV files in your browser
Drop a CSV and read it as a clean, sortable, searchable table. The file never leaves your device — parsing runs entirely in your browser.
Click to upload or drag and drop
Drop your CSV, Excel, or JSON file
Max 10MB
Runs 100% in your browser — this page makes no network request with your data.
- 1
Drop a .csv (or .tsv) file into the box — no signup, no upload.
- 2
Sort any column, search across rows, and page through large files.
- 3
Need analysis, not just viewing? Send the file to the workspace with one click.
Worked example: checking a 12MB subscriber export before import
A lifecycle marketer exports 40,000 contacts from Mailchimp as a CSV and needs to confirm the file is clean before importing it into a new CRM. Opening it in Excel would risk mangling the phone numbers and the zip codes with leading zeros, so instead they drop the file straight onto the CSV viewer.
The whole table renders in the browser. They sort by the signup-date column to see the newest contacts, search for a specific domain to spot test accounts, and page through to confirm the header row and column count match what the CRM import expects. Because parsing runs locally, the subscriber list never leaves their laptop, which matters when the data includes personal email addresses.
Once the file checks out, they click "Turn this into a report" to send it to the workspace, where they can chart signups by source and build a shareable summary for the team.
A CSV viewer opens a comma-separated values file and renders it as a readable table of rows, columns, and headers, without importing it into a spreadsheet program that might quietly rewrite your data. To open a CSV online, drop it into a browser-based viewer like the one above: it parses the file on your own machine and shows a sortable, searchable grid in seconds, with nothing uploaded.
That is often all you actually need, to read a file, confirm its columns, or check that an export looks right before it moves on to a CRM, a database, or a teammate. Viewing is not editing and not analysis, so the right tool depends on which of the three you are doing.
Every way to open a CSV file, compared
You have five realistic ways to open a CSV, and the right one depends on whether you want to read it, edit it, or run numbers on it. Reading is the most common need and the one people over-engineer, reaching for Excel when a viewer would be faster and safer.
The honest tradeoff is that the tools built for editing and calculation are exactly the ones most likely to alter your file on the way in. A viewer that only displays the data cannot corrupt it.
| Method | Best for | Watch out for |
|---|---|---|
| Browser CSV viewer (this page) | Reading or checking any CSV fast, no install, fully private | View-only; very large files should be split first |
| Microsoft Excel | Editing and formulas in a familiar grid | Silently mangles leading zeros, dates, and long IDs; 1,048,576-row ceiling |
| Google Sheets | Sharing and light collaboration | 10-million-cell cap per file; same reformatting risks as Excel on import |
| Text editor (VS Code, Notepad++) | Seeing the raw delimiters, quoting, and encoding | No table view; wide files are painful to read |
| Command line (head, less, column, csvkit) | Peeking at huge files and scripting | Requires terminal comfort; no point-and-click sorting |
Why Excel mangles CSV files (and how to open one cleanly)
Excel treats a CSV as raw text and guesses the data type of every column the moment you open it, so it rewrites values it decides are numbers or dates, even if you never touch a cell. Because a CSV carries no formatting metadata, Excel's guess is final: save the file and the original characters are gone.
According to Microsoft's own specifications, a worksheet also tops out at 1,048,576 rows by 16,384 columns. Open a CSV with more rows than that and Excel loads the first million-odd and silently drops the rest, with no warning that data is missing. These are the specific ways a clean export gets corrupted:
How Excel rewrites CSV values on open
- Leading zeros vanish: 02116 becomes 2116, breaking ZIP codes, zero-padded SKUs, and account numbers.
- Long numbers go scientific: a 12-digit order ID like 105000000001 is displayed as 1.05E+11 and the exact digits are lost.
- Text that looks like a date gets rewritten into Excel's date format, permanently.
- Encoding is guessed: a UTF-8 file saved without a byte-order mark is often read as Windows-1252, turning accented characters into garbled sequences.
- Rows beyond 1,048,576 are dropped silently, so a truncated file looks complete.
How to open a CSV without the damage
Do not double-click a CSV to open it in Excel. Use Data, then From Text/CSV, and set the affected columns to Text in the import step so the zeros and IDs survive. Simpler still, read it in a viewer that never reformats anything, then open it in a real CSV editor only when you genuinely need to change values. Both keep every character exactly as written.
Delimiter and encoding problems, and how to fix them
If a CSV opens as one giant column, or shows garbled characters where accents should be, the delimiter or the encoding is the culprit, not a broken file. These are the four issues that account for almost every unreadable CSV.
- Semicolons instead of commas: exports from European locales often use a semicolon delimiter because the comma is their decimal separator. A viewer that auto-detects the delimiter reads them fine; Excel keys off your regional settings and may not.
- Tabs instead of commas: a tab-separated file (.tsv) is the same idea with a different separator. This viewer handles TSV directly, so you do not have to rename or convert anything.
- Quoted fields: a value that itself contains a comma or a line break is wrapped in double quotes per the CSV convention (RFC 4180). Splitting naively on commas shreds these rows; a real parser respects the quotes.
- UTF-8 and the BOM: a byte-order mark is three invisible bytes at the start of a file. When it is missing, some tools misread UTF-8 and produce mojibake (cafe turns into cafA-with-symbols); when it is present, a naive parser can show stray characters before your first header.
The phantom sep= row
Some programs prepend a line like sep=; to a CSV so Excel knows the delimiter. Excel reads and hides that line, but other parsers, and some viewers, show it as a strange first row above your headers. If you see it, delete that top line and the file reads normally everywhere.
Opening large CSV files without crashing your browser
For files past roughly 10MB or 50,000 rows, stop trying to load the whole thing into a browser tab or Excel and switch strategy. This viewer handles files up to about that size smoothly; beyond it, the tab slows down because every row becomes a DOM element.
When a file is genuinely large, work with a slice or a different engine rather than fighting the memory limit head-on.
- Split first: break the file into parts with the CSV splitter, which keeps the header on every part and also runs in your browser.
- Peek from the command line: head -n 100 file.csv shows the top rows, and wc -l file.csv gives the exact line count (subtract one for the header) without opening anything.
- Know the Sheets ceiling: Google Sheets caps a spreadsheet at 10 million cells, which is only about 385,000 rows once you have 26 columns, so a wide export hits the wall sooner than you expect.
- Analyze instead of read: when the goal is to compute over the whole file rather than eyeball it, send it to the workspace, where real Python runs over the full dataset regardless of row count.
Frequently Asked Questions
Everything you need to know about using AnalyzeData.
Yes — the file is parsed by JavaScript in your browser and never transmitted. You can verify this in your browser's network tab: no request carries your data.
Files up to about 10MB / 50,000 rows open smoothly in most browsers. Beyond that, split the file first — our CSV splitter handles that, also in-browser.
That is exactly what this page is for: a fast table view without installing anything or fighting Excel's type mangling (leading zeros, dates, long numbers all display as-is).
Yes. The viewer runs entirely in your browser, so there is nothing to install and no admin rights to request; it works the same on a locked-down work laptop or a Chromebook. Your file is parsed by JavaScript on your own machine and never uploaded, which is often exactly what makes it usable where IT blocks desktop apps.
Drop the file here and read it as a table first. Sort a column to surface blank or malformed values, search for test rows, and confirm the header names and column count match what the destination tool expects. Catching a shifted column or a stray delimiter here saves a failed import later, and nothing uploads while you inspect it.
The delimiter does not match what the program expects. Your file probably uses semicolons or tabs while the tool assumed commas, so every row is treated as one long value. Open it in a viewer that auto-detects the delimiter, or in a text editor to confirm which character separates the fields, then tell your spreadsheet program to use that delimiter on import.
On the command line, wc -l file.csv returns the number of lines; subtract one for the header row to get the record count. In a browser viewer, the row count is shown once the file is parsed. Counting lines this way is also the fastest way to confirm Excel did not silently truncate a large file at 1,048,576 rows.
That is an encoding mismatch. Stray characters before the first header usually come from a UTF-8 byte-order mark being read by a tool that does not expect one. Garbled accented letters mean a UTF-8 file is being read as Windows-1252. Re-open the file specifying UTF-8 encoding, or use a viewer that detects it automatically, and the characters render correctly.
Yes. A browser-based viewer runs in any modern mobile browser, so you can open and read a CSV on a phone or tablet without an app or an Office license. Because parsing happens on the device, the file is never uploaded, which is useful when someone emails you an export and you just need to check it on the go.
When a table isn't enough
The workspace runs verified AI analysis on the same file and turns the results into a report you can send.
Analyze this data instead