Choose a worksheet and a format
Open an XLSX file, then choose a worksheet. The first row becomes the field names. Empty and duplicate header cells receive unique names so they cannot overwrite another column. Each following row becomes a JSON object, or a row in CSV/TSV. Trailing cells that contain only formatting are ignored. Empty rows and columns within the data range are retained and count toward the export limits.
Create Excel from JSON
Switch direction, paste a JSON array of objects, and create the workbook. The union of object keys becomes the header. Missing fields become empty cells, booleans remain booleans, and nested objects or arrays become JSON text cells. Text beginning with an equals sign stays text; it is never turned into a formula.
[
{"account_id": "9007199254740993", "name": "Ada", "active": true},
{"account_id": "2", "name": "Grace", "active": false}
]CSV and type conversion
CSV and TSV input remain text when exported to Excel, preserving leading zeros and avoiding date guesses. On export from Excel, formula protection adds an apostrophe to formula-like cells. This changes their textual representation; turn it off when a program needs the original values.
Need to inspect delimited data first? Use the CSV viewer or TSV converter.
Frequently asked questions
Which Excel formats can I open?
XLSX workbooks. Legacy binary XLS, macro-enabled XLSM and password-encrypted workbooks are not supported. Save a copy as XLSX before opening it.
What happens to formulas and formatting?
Formulas are not executed. Cached formula results are used, and missing cached results become empty cells. Formatting, charts and macros are not included in the text export. Excel dates become ISO date strings.
How do you preserve long numeric IDs?
When creating XLSX from JSON, numbers with more than 15 significant digits are stored as text, because Excel would round them. Numbers already rounded in an uploaded workbook cannot be recovered; store exact IDs as text in the original file.
How large can a workbook be?
Up to 20 MB, with at most 50,000 data rows per sheet, 1,000 columns and 250,000 cells in total. Processing runs in a cancellable worker. JSON/CSV input for workbook creation is limited to 10 MB.