How to Fix Mixed Date Formats in a CSV (9/11/2026: September or November?)

A date column that mixes 9/11/2026, 2026-10-05, Nov 3, 2026 and an empty cell can't be sorted, filtered or imported reliably, and the worst entries are the ones that look fine. 9/11/2026 is 9 November if the person who made the file writes day first, and 11 September if they write month first. Both are valid. A tool that guesses wrong shifts the date by up to a few months and says nothing.

The free CSV date format converter reads a date column, works out the order from the data where it can, rewrites every date to the format you choose and lists the cells it couldn't read. This post covers the logic, so you can check any converter's output and clean the column by hand if you need to.

Convert a date column →

Why the same list has five date formats

Lists get assembled from several places, and each leaves its own habit in the column:

• A US form tool writes 11/09/2026 for 9 November.
• A UK or European export writes 09/11/2026 for the same day.
• A scraper writes "Nov 9, 2026" or "9th November 2026".
• A database export writes 2026-11-09, sometimes with a time such as 2026-11-09T14:30:00Z.
• A spreadsheet that opened the file fills in its own idea of what a date is.

When lists get merged (merging lead lists without duplicates covers that), the column ends up holding all of them. Signup dates, last-activity dates and purchase dates are the usual victims, and they often drive segments: "contacted in the last 90 days", "signed up this year".

The ambiguity problem, and how to settle it

A cell like 9/11/2026 has two numbers that could be day or month. The converter's options for this are "work it out", "day first" and "month first". Working it out uses the same rule you can apply by eye:

1. Scan the column for a first number above 12. 13/11/2026 or 27/03/2026 can only be day-first. If you find one, and no value with a second number above 12, the whole column is day-first.
2. Scan for a second number above 12. 11/13/2026 can only be month-first.
3. If both appear, the column is mixed, and you need to split it by source before converting.
4. If neither appears, the data alone can't tell you. Ask where the file came from. If you can't find out, pick the likelier order for that source and write the choice down. A file of purely US signups is month-first; one exported from a European CRM is day-first.

Dates written with a month name (Nov 3, 2026, 3 November 2026) and four-digit-year-first dates (2026-10-05) are never ambiguous, so they convert directly.

Pick one output format

The converter offers four ways to write a date: 2026-11-09 (year-month-day), 11/09/2026 (month/day/year), 09/11/2026 (day/month/year) and 9 November 2026. Pick year-month-day unless the destination requires something else:

• It is only ever read one way.
• A text sort puts it in date order.
• Most importers and databases accept it.

Use the long form only for a document people read. Use the numeric forms only when a destination insists on them, and then convert as the very last step so no later tool reads them the wrong way round.

Impossible dates and blanks

Two kinds of cell can't be converted, and a good converter reports them instead of hiding them:

• Impossible dates. 31 Feb 2026 or 2026-13-01 aren't dates. The converter's sample includes exactly this case. Rolling 31 Feb into early March would invent a date, so the cell is listed as unreadable for you to fix at the source or clear.
• Blank cells. An empty cell stays empty. Whether to fill it is a business decision (a missing signup date shouldn't become today's date).

Read the unreadable list before you import. If it is five rows, fix them. If it is two thirds of the file, the column isn't a date column, or the file has shifted columns; the CSV misplaced data finder covers that case.

The by-hand method in a spreadsheet

If you prefer to do it yourself:

1. Format the date column as text before pasting, so the sheet doesn't reinterpret it.
2. Add a helper column that splits the cell on the slash or dash.
3. Write one formula per source order, and apply each only to the rows from that source.
4. Wrap the result in a text function that outputs year-month-day.
5. Copy and paste as values, then delete the helper.

It works, and the weak point is step 3: one formula applied to a mixed column silently misreads the rows from the other source. The tool's value is mainly the whole-column check and the unreadable list.

What the date fix doesn't cover

Cleaning the date column changes how the file is read. It doesn't touch whether the people on it can be reached. A list with perfect dates and a block of dead mailboxes still hurts your sending reputation on the first send. Sort that out before the file goes anywhere:

1. Fix the dates, and any other columns that need cleaning.
2. Remove duplicates and rows with no email address.
3. Verify the emails at /verify. Valid rows go in. Invalid rows are suppressed. Catch-all and risky rows go in a smaller, slower group; what to do with catch-all emails explains the call.
4. Import. The guides for HubSpot and Salesforce list what each expects in a file.

The converter hands the addresses in your file straight to the verifier when you finish, so step 3 doesn't mean re-uploading from scratch.

Quick answers

• Is 2026-11-09 safe in every tool? It is the least ambiguous format, but check the destination's own import guide for a format requirement.
• Does a time on the end matter? The converter reads dates with a trailing time and keeps the date. If you need the time too, keep a separate column.
• Are two-digit years handled? Two-digit years are read as 20xx. Check old files, where a 98 should be 1998.
• Will it change my other columns? No. Only the column you choose is rewritten.

Try it on your own list

Upload a sample at /verify. The first 100 checks are free with a Google sign-in, no card.

Pro, $39 a month: Unlimited email verification.* Never metered, never cut off. See pricing.

*Fair use: the first 100,000 checks a month at full speed, then slower, never cut off.