Preparation node
Formula
Add or replace columns with SQL expressions.
What it does
Adds new columns, or replaces existing ones, using SQL expressions evaluated a row at a time. Each row of the node is one output column name plus the expression that fills it, and you can add as many as you need. Expressions are checked as you type, so a mistake is caught in the builder rather than at run time.
Inputs and outputs
| Inputs | One. |
| Outputs | One. |
When to use it
- Compute a new value: a total, a ratio, a flag
- Clean or reformat a column in place
- Sort rows into categories with CASE WHEN
Examples
Arithmetic
[price] * [quantity]
Output column: total. The bread-and-butter case.
Text cleanup
trim(upper([code]))
Reuse the same column name to overwrite it in place.
Category
CASE WHEN [score] >= 90 THEN 'A' WHEN [score] >= 80 THEN 'B' ELSE 'C' END
CASE WHEN is how conditionals are written here; there is no IF().
Fallback value
coalesce([nickname], [first_name], 'Friend')
Takes the first value that isn't null.
Build on the row above
1: subtotal = [price] * [qty] 2: total = [subtotal] * 1.2
Rows run in order, so a later one can use a column an earlier one just made.
Tips
- Rows are evaluated top to bottom, and a later row CAN reference a column an earlier row created. That's how you build a value up in readable steps instead of one long expression.
- Wrap column names in [brackets]. Names match exactly, capitals included.
- Reuse an existing column name as the output to overwrite that column in place.
- Use the Columns button to insert a name correctly, and Syntax help for the function list.
- Dates are written as plain ISO text ('2026-08-26') and compared as real dates. There is no locale guessing, deliberately.