Excel swaps CSV month and day: convert from the source date order
By ON Lab · Sources checked November 8, 2026
If Excel interpreted month/day text as day/month, changing the cell's display format does not recover the intended date. Return to the original CSV text, establish its source convention and import or convert that text with the matching date locale. Verify the year, month and day before using the result in calculations.
Historical readers reported both silently swapped dates and mixed date/text columns from CSV input. One explicitly distinguished an incorrect underlying date from an unwanted display style. We checked Microsoft's current import and Power Query data-type guidance. We did not run Excel or Power Query, and the small date comparison below is a Python illustration of ambiguity rather than an Excel reproduction.
Preserve the original text before conversion
Keep the source CSV unchanged and work in a new workbook or query. Inspect a few original values in a plain-text viewer, along with the exporter's documented date convention. A slash-separated value such as 03/01/2012 does not reveal by itself whether March 1 or January 3 was intended.
Use a known source record or an unambiguous date to establish the convention. Do not infer it solely from your computer's current region or from the date format you prefer to display. A source may also contain mixed conventions, in which case choosing one locale for the whole column is not enough.
Why only some rows appear broken
| Source text | Month/day reading | Day/month reading |
|---|---|---|
| 03/01/2012 | March 1, 2012 | January 3, 2012 |
| 03/13/2012 | March 13, 2012 | Invalid month 13 |
We checked these interpretations with explicit Python date formats. The example explains how an ambiguous row can silently become another valid date while a different row fails under the same assumed convention. It does not assert exactly how every Excel version will store a failed conversion.
Changing a display pattern only changes how an already interpreted date is shown. If March and January were already interchanged during parsing, the source meaning must be restored through a correct conversion, not hidden under another number format.
Use a documented import route
Microsoft's Excel guidance distinguishes opening a CSV directly from importing text with control over column interpretation. In a desktop version with the documented route, start from a blank workbook and use Data → From Text/CSV. Inspect the preview before loading. Choose the transformation route when you need to control the date conversion.
Power Query can detect and apply types automatically. Read the applied steps: a previous type conversion may already have interpreted the source text. To choose the correct conversion, work from the original text stage rather than converting an already wrong date back to text and treating that as the original input.
The current Power Query documentation describes selecting a column, opening Change type → Using locale, then choosing the intended data type and locale. Select a date type and a locale whose convention matches the source text. For genuinely US month/day input, an appropriate US English locale expresses that interpretation; a different source convention needs its corresponding choice.
These are source-based steps, not verified screenshots of our own Excel session. The available import and transformation interface depends on the Excel edition and platform. If you are using a web workflow without this editor, do not assume the desktop path is available. Use the documented route supported by your application or preserve the column as text until a suitable conversion is possible.
Check the date components, not only the appearance
- Retain a copy of the original text column for comparison.
- Choose a known ambiguous row and a known unambiguous row.
- Inspect the converted year, month and day separately.
- Check conversion errors and unexpected null values across the column.
- Only then load or use the dates in calculations, filters or reports.
Microsoft's Power Query guidance describes extracting date parts as a way to verify interpretation. A successful conversion without errors is insufficient when both possible date orders produce valid dates. Check a source record whose intended month and day you actually know.
If the source uses two-digit years, extra spaces, mixed separators or a time component, keep those details in the investigation. Do not silently invent a century or discard time information merely to make a date-only example pass. The original historical examples included details that a real import must review.
Keep other CSV problems separate
CSV character decoding, identifier preservation and date interpretation are different decisions. The CSV encoding guide covers garbled characters. The leading-zero guide covers identifiers that should remain text. Selecting the correct encoding does not tell Excel whether 03/01 means March 1.
If the original file has already been overwritten after an incorrect conversion, a formatting change cannot prove recovery of the original meaning. Retrieve an unchanged export or another reliable source record and compare it. This guide does not claim to reconstruct lost date conventions from ambiguous numbers alone.
Checked November 8, 2026. Microsoft procedures were reviewed, but Excel UI, Power Query execution and the readers' files were not tested. The local illustration only verifies two explicit date interpretations. Historical questions do not establish current search volume or a version-specific repair.