Working with Excel spreadsheets in VisiData isn’t much different than working with CSV files or other tabular data, but when you open an Excel file, the first thing you’ll see is a listing of the sheets it contains. (Navigate to the sheet you want to explore, and then press Enter to open it.)

## How to fix funky spreadsheets’ column names¶

By default, VisiData considers the first row of a tabular dataset to be its columns’ names. That’s a safe assumption for many formats, but Excel spreadsheets often buck that trend, with titles, notes, or other cruft coming before the actual column names.

As previously noted, you can manually edit a column’s name manually by pressing ^, or by editing it from the Columns Sheet. But you can also auto-populate all unnamed column names at once by doing the following:

• Navigate to the row that contains your desired column names
• Type g^ to hoist the names into the column header

If you’d like to have VisiData not to load the first row as the header, you can do one of the following:

• Pass the --header 0 option when you launch vd from the command line.

• From within VisiData, do this:

• Press O to open the Options Sheet
• Set the header option cell to 0