A 35 MB spreadsheet is only 6 MB
of data. The rest is bloat.
Unreadable content or won't open? Rescue it with the repair tool →
standbyChoose a file
The processing server is unreachable right now (maintenance or network). Uploads will fail; please try again later.
Unreadable content or won't open? Rescue it with the repair tool →
standbyChoose a file
Usually tens of thousands of duplicate styles pasted in from elsewhere, conditional formats and data validation fragmented by dragging, invisible formatting on empty rows and columns, pivot caches holding a copy of the source data, high-resolution images and leftover shared-workbook revisions. The actual data is often a small part.
No. After every tier the output is compared with the original cell by cell (values, formulas, formats, conditional formats, data validation) and re-checked with openpyxl. Any mismatch drops to a gentler tier; if all three fail, no file is produced.
That's usually formulas: volatile functions (OFFSET, INDIRECT, TODAY) or whole-column references make every edit recalculate a lot. Try the formula optimizer, or run a check-up to see where the problem is.
xlsx, xlsm, xltx and xltm up to 100 MB each, macros kept. Pick several files to process them in one go and download a zip. Files up to 30 MB can also be processed entirely in your browser.