Unpivot Transform

Unpivot moves selected columns from a wide dataset into rows, producing long-format data. This operation is also known as melting.

Basic Usage

To unpivot selected columns:

1. Select Unpivot from the transform menu.
2. In Columns to Keep, select at least one identifier column.
3. In Columns to Unpivot, select at least one column to turn into rows.

Configuration Options

Basic Options

• Columns to Keep: Select at least one identifier column. These columns remain unchanged in the output.
• Columns to Unpivot: Select at least one column to turn into rows. Leaving this empty does not select all remaining columns, and you cannot apply the form until you make a selection.

Output Columns

• variable: Holds the original names of the unpivoted columns.
• value: Holds the corresponding values from those columns.

For nodes configured in the current app, these names are fixed. Older or API-created nodes may keep custom output names, but the current Unpivot form cannot edit them.

Examples

Example: Unpivoting Monthly Sales Data

Input Dataset:

Product	Jan	Feb	Mar
Laptop	1000	1200	900
Smartphone	1500	1600	1400
Tablet	500	450	550

Configuration:

• Columns to Keep: Product
• Columns to Unpivot: Jan, Feb, Mar

Result:

Product	variable	value
Laptop	Jan	1000
Laptop	Feb	1200
Laptop	Mar	900
Smartphone	Jan	1500
Smartphone	Feb	1600
Smartphone	Mar	1400
Tablet	Jan	500
Tablet	Feb	450
Tablet	Mar	550

tip

Long format works well for time series, grouped analysis, and visualizations that expect one observation per row.

caution

Unpivoting can multiply the number of rows. Estimate the output size before applying it to a large dataset.

Best Practices

1. Choose Columns to Keep that clearly identify each original record.
2. Missing values in unpivoted columns remain missing in the long result.
3. If the selected columns contain mixed types, you may need a Convert Column Type step after unpivoting.
4. Preview variable and value before using them downstream.