Excel Broke the Phone Numbers in My CSV

Updated July 2026 · By the CSVUndo team · 6 min read

The quick answer: a phone column is the one place Excel makes three of its worst guesses at once. It drops the leading 0 from 07911123456, strips the + from +441234567890, and squashes a long number into 4.47E+11. All three happen because a CSV is plain text and Excel decides each cell’s type for you — and it reads a phone number as a quantity. The fix is to import the phone column as Text (Data → From Text/CSV, set the column’s Data Type to Text) so Excel never parses it. Below: exactly why each symptom happens, an honest table of what you can and can’t recover after a save, and the click-by-click import.

The three symptoms at a glance

Every broken phone column is one of the four rows below. The right-hand column is the honest part — whether the damage can be undone after the file has been saved back to CSV.

Why Excel does it

A CSV file is plain text. There is no place in it to write “this column is phone numbers, treat every character literally,” so when Excel opens the file it guesses the type of each cell — and a run of digits looks like a number. That single guess is the root cause of all three symptoms:

The leading 0. A UK mobile such as 07911123456 is stored as the number 7911123456, because the value zero-followed-by-digits as a number is just the digits. The zero was never wrong in your file; it was removed the moment Excel decided the cell was numeric. This is the same mechanism that eats ZIP codes and SKUs, covered in our leading-zeros fixer.

The + sign. A leading + is how Excel begins a positive number or, historically, a formula, so it treats +441234567890 as the numeric value and discards the plus, usually leaving 441234567890. The dialling + is a character to a human and a sign to Excel, and Excel wins.

The 4.47E+11. Once a cell is a number, Excel shows anything with 11 or more digits in scientific notation to fit the column, so 447911123456 appears as 4.47E+11. At 15 digits or fewer this is display only — the digits are still stored. But Excel keeps at most 15 significant digits, so a 16-or-more-digit value has its final digits replaced with zeros on the way in; Microsoft documents this 15-digit limit directly. See our scientific-notation fixer for the display side of this.

Import the column as text

The one reliable prevention is to stop Excel parsing the column at all — bring it in as Text. Double-clicking the CSV does the damage before you can object, so use the import route instead:

1. Open Excel with a blank workbook. Don’t double-click the CSV file.
2. Go to Data → From Text/CSV, select your CSV, and click Import.
3. In the preview, click the phone column’s header (use Transform Data if you need the full editor) and set its Data Type to Text.
4. Click Close & Load. The +, the leading 0 and every digit arrive exactly as written.

The same idea applies if you are building the file yourself: protect the phone column as text before you save, not after. A value stored as text will sit left-aligned and may show Excel’s green “number stored as text” triangle — that warning is harmless here and is exactly what you want for a phone column. We cover it in number stored as text in Excel, and the wider “open without breaking” routine in open a CSV in Excel without breaking it.

What you can recover after a save

This is where honesty matters, because a saved CSV has already discarded whatever Excel dropped. Recovery depends entirely on which symptom you had:

A dropped leading 0 can be re-added, but only if you know every number should be the same width. A UK mobile is 11 digits, so =TEXT(A2,"00000000000") re-pads 7911123456 back to 07911123456. If the column mixes landlines, mobiles and international numbers you cannot safely guess the width, and re-padding will invent wrong zeros.

A stripped + can be re-added only if you know every value is international — then prefixing + is safe. If some rows were national and some international, the + is genuinely gone, because nothing in the saved file records which rows had it.

Scientific notation expands cleanly whenever 15 or fewer digits were kept: widen the cell, or better, re-import the column as text and the full digit string returns. Our scientific-notation tool does this on the raw CSV.

A 16-or-more-digit number that was truncated is the one unrecoverable case. Once Excel replaced the tail with zeros and you saved, those digits are stored nowhere in the file. No formula or tool can rebuild them — the full story is in Excel changed my long number’s last digits to zeros. You must re-export from the original source; better still, import as text next time so it never happens.

Frequently asked questions

Sources: community discussion of keeping the plus sign when importing a CSV; Microsoft’s note that the last digits change to zeros past 15 significant figures.