Save

The Save node writes one row to a table. If the table has a key, whichever of your mapped values match its key columns are what identifies the row, so the same visitor writes the same row again rather than creating a duplicate. Keys ignore capitalisation and surrounding spaces, so [email protected] and [email protected] are the same row.

How It Works

1. Pick a table
2. Map each value you want to store to a column name (a name that isn't a column on this table is ignored). If the table has a key, include a value for each key column here too
3. Choose what happens when a row for that key already exists

When A Row Already Exists

• Create or update (default): the existing row is updated. Only the columns you mapped are written, so values this step didn't collect are left as they were, and a column you mapped to an empty value is cleared
• Create only: the existing row is left untouched
• Update only: nothing is written, since there is nothing to update

A Delete node removes a row a flow can find. Deleting in bulk stays a portal operation.

Branches

Skipped is not an error. It's how Create only and Update only tell you they deliberately left an existing (or missing) row alone, the same way Look up's Not found tells a returning visitor apart from a new one. Saved means a row was genuinely written.

Choosing Columns

Each value is addressed to a column you pick, not to a name you type, so renaming that column later leaves this step working.

If a Row Will Not Fit

A row holds about 8,000 characters across all its columns, counted on the values themselves whether or not the column is encrypted. A row over that is not saved at all, and nothing is trimmed to make it fit, so a row you already had is left as it was. The step takes Error.

Output Variable

{{variable}}.saved
{{variable}}.created