Google Sheets says it detected the delimiter automatically, why is it still wrong?

Google Sheets' automatic delimiter detection samples the file and picks the most likely separator, but it can guess wrong on files with an unusual or mixed delimiter, or where the real delimiter appears rarely relative to other punctuation in the data. Selecting the delimiter manually during import overrides the guess.

I fixed the delimiter but a handful of rows still have shifted columns, why?

That's a second, independent problem: an unescaped delimiter or a stray quote inside one specific field, not the file-wide delimiter setting. Fixing the import delimiter solves it for every row except the ones where a value itself contains that character without being quoted.

Does re-importing overwrite my existing sheet?

Using File > Import with the option to "Replace current sheet" or "Insert new sheet" won't touch other tabs in the same spreadsheet, but it's safer to import into a new sheet first, verify it looks right, then delete the broken one.

Why does IMPORTDATA give me the same one-column problem as a normal import?

IMPORTDATA has no delimiter selector at all, it assumes comma-separated data unconditionally. If the source file uses a different delimiter, download it locally and use File > Import with an explicit separator instead, or wrap the formula in SPLIT() on the correct character.

Can I prevent this from happening on every file a teammate sends me?

Not on your end alone, the fix belongs in how the file is generated. Ask whoever exports it to standardize on one delimiter and quote any text field that might contain it, or run each file through a free syntax check like CSV Lint before it's distributed.

Fix: Google Sheets CSV Columns Misaligned

Two different symptoms get reported as "too many columns" in Google Sheets: everything piles into column A, or most columns split correctly but a few rows shift partway through. They have different causes, here's how to tell which one you have and fix it for free.

Symptom A: the entire file is in column A

This is a delimiter mismatch, Google Sheets' auto-detection guessed comma but your file uses semicolons, tabs, or a pipe (|), or vice versa.

Fix, if you haven't imported yet:

File → Import → Upload, select the file.
In the import dialog, find Separator type and change it from "Detect automatically" to the actual delimiter (Comma, Semicolon, Tab, or Custom).
Choose Replace/Append/New sheet, then Import.

Fix, if it already imported wrong:

Select column A.
Data → Split text to columns.
A separator dropdown appears at the bottom-right of the selection, pick the actual delimiter (or "Custom" to type it in).

Symptom B: most columns are correct, but a few rows shift

This is a per-row problem, not a file-wide delimiter issue: a specific field contains the delimiter character (usually a comma inside an address, name, or description) without being wrapped in double quotes. Google Sheets has no way to know that comma is part of the value rather than a column boundary.

Find the affected rows with our free CSV Lint tool, paste the file and it flags every line whose field count doesn't match the header, so you're not scrolling through thousands of rows manually.

If you imported with IMPORTDATA instead of File → Import

=IMPORTDATA(url) pulls a CSV from a URL directly into a cell formula rather than through the import wizard, and it has no delimiter picker at all, it assumes comma. If the source is semicolon- or tab-delimited, the whole row lands in one cell the same way it would from a wrong auto-detection, and there is no dialog to override it.

Fix: download the file locally first and use File → Import with an explicit separator instead of IMPORTDATA, or wrap the formula's result with SPLIT() on a fixed delimiter, e.g. =ARRAYFORMULA(SPLIT(IMPORTDATA(url), ";")), though this only works cleanly when no field itself contains that delimiter.

Preventing this the next time you export the file yourself

If you control how the CSV is generated in the first place (an export script, a database query, another spreadsheet), the fix belongs upstream instead of being repeated on every import:

Pick one delimiter and standardize on it. Comma is the most widely assumed default across import tools; if your locale or source system defaults to semicolon, say so explicitly to whoever receives the file instead of leaving it to be guessed.
Always quote text fields that could contain the delimiter (addresses, names with suffixes, free-text notes). Most CSV libraries (Python's csv module, pandas' to_csv) support a QUOTE_NONNUMERIC or QUOTE_ALL mode that wraps every text field automatically instead of only when a delimiter happens to be present.
Run a quick check before sending it out. Paste the file into CSV Lint once before distributing it, catching a field-count mismatch before a colleague spends time debugging their import is faster than fixing it after the fact.

Doing this in How To CSV

Auto Fix detects the real delimiter and normalizes quoting for fields that contain it, so the re-imported file lands correctly on the first try instead of needing a manual Text-to-Columns pass.

Fix it before you re-import, free

Normalize delimiters and quoting in your browser, no upload.

Fix My File

Turn this into a saved workflow

Create a free account to save the steps from this guide as a reusable workflow and re-run it on any file, from any device.

Follow HowToCSV on Google

Add us as a preferred source on Google Search so our latest CSV guides and tutorials surface more often in your Top stories.

Add HowToCSV as a preferred source