Branch the workflow: route to then or else based on a condition
Branches the whole workflow with a no-code condition builder. If the condition holds, the flow takes the green then path; otherwise the amber else path. It routes the flow — it does not filter rows; both branches receive the same, unchanged dataset, and nodes reachable only through the untaken branch are skipped. Check either the row values (field / operator / value, matching any row, every row, or the first row) or the dataset's row count.
| Field | Description |
|---|---|
| Input variablerequired | Which variable to use as the upstream dataset to read. |
| Checkrequired | What to test: the row values (field/operator/value conditions) or the dataset's row count. |
| Rows to check | Rows mode: take the then branch when any row matches, every row matches, or the first row matches. Note: with an empty dataset, 'every row' is vacuously true, while 'any'/'first' are false. |
| Conditionsrequired | One or more field/operator/value rows (same operators as CSV Filter). With several rows, combine them with all (AND) or any (OR). |
| Row-count check | Dataset mode: compare the dataset's row count against a number (e.g. row count > 0). No rows are read. |
Open the node's real configuration, run it on the sample below, and see the result. Nothing is saved.
Sample input · data
| name | age | city |
|---|---|---|
| Alice | 34 | Tunis |
| Bob | 28 | Sfax |
| Cara | 41 | Tunis |
| Dan | 23 | Sousse |
| Eve | 37 | Sfax |
This is a live preview — the sample and results are temporary and never saved.
It routes the whole flow, not individual rows — to keep or drop rows, use CSV Filter instead.
A referenced column that doesn't exist fails the node with a clear error, instead of silently taking the else branch.