Excel File Is Corrupt and Cannot Be Opened — How to Repair .xlsx Without Losing Data

Excel refuses to open a workbook with “the file is corrupt and cannot be opened” for one dominant reason, and it is less mysterious than it sounds. An .xlsx file is not a single document blob — it is a ZIP archive holding a set of editable XML files: one per worksheet, one for shared strings, one for styles, one for the workbook index. Excel is extraordinarily strict about the ZIP container: if the archive index at the tail of the file has been damaged — a download that stopped halfway, a cloud sync interrupted, a USB stick unplugged mid-write, an app that crashed while saving — Excel throws this error and refuses the whole file, even though the XML worksheets inside are almost always intact.

Which means: the workbook with your quarterly figures, your VAT register, your course grades is probably sitting there unharmed inside the broken container. The fix is to rebuild the container — and that can be done without opening Excel, without installing anything, entirely in your browser:

Embedded — usable right here: Repair a corrupted Excel file (full page)

Drop the damaged file here

or click to choose — XLSX,XLSM,ZIP · rebuilt on your device, never uploaded

Rebuilding in your browser…

Scanning entries and healing XML — nothing uploaded

The embedded tool rescans the raw bytes of the damaged workbook for readable ZIP entries (it does not trust the broken index), recovers every sheet, shared strings and formatting that survived, repairs truncated XML, and writes out a clean .xlsx that Excel accepts. The report after repair tells you exactly which parts were healed and which were physically missing — no pretending about data that isn’t in the file.

Things worth trying before (or alongside) a repair

These built-in routes cost nothing and occasionally help:

  1. File → Open → select the file → Open and Repair. Excel’s own recovery. It works when the XML is clean and only the container flag is confused; it usually drops charts and some formatting when it does work, so compare against the embed above.
  2. Check the file size. If a colleague emailed you the workbook, ask them for the original file size. A file that arrived 30KB short of the original is truncated, and a re-download is the true fix.
  3. Try protected-view off. Sometimes Excel conflates “blocked file, downloaded from the internet” with corruption. Right-click → Properties → Unblock → reopen. Costs nothing.
  4. Google Sheets / LibreOffice import. Their parsers are more forgiving of odd containers. If another app can open it, immediately re-save as a fresh .xlsx and you may be done.

If none of those land, the container rebuild above is the reliable route — it does not depend on Excel co-operating.

Why “corrupt” is usually the wrong word

Half the time the workbook itself is fine; the ZIP index is what Excel rejects. That is why the error appears after a transfer but not on the machine where the file was created — the creation machine holds the file whose index was written before the interruption happened elsewhere. Practically: ask whoever sent you the file to re-share it from their original copy, and in parallel run the repair above — whichever succeeds first wins, and neither costs you an upload of your data.

Frequently asked questions

Will Open and Repair in Excel recover everything?
It recovers values when the XML parts are intact, but it often strips formatting, drops charts and rewrites formulas it cannot parse. A container rebuild keeps the recovered parts exactly as they are stored — formulas, styles and all sheets.
Is it safe to send my workbook to an online repair service?
The embedded repair below never asks you to send anything: the file is parsed inside your browser tab, so payroll data and client sheets never leave your device. If a service requests the upload, that is precisely the data you should be least willing to hand over.
Why did Excel only complain after a transfer?
Because .xlsx is a ZIP container and the index that validates it is written at the end of the file — the part most likely to be lost when a sync, copy or download truncates.