Data Tables

Data Tables are a built-in database for workflows. Unlike node outputs (which only exist within a single run), data in a table persists indefinitely and is accessible by any workflow in your workspace.

Use Cases

• State tracking: remember which items have been processed, track run counts, store checkpoints
• Accumulating results: collect one row per execution across many runs, then query the accumulated data
• Cross-workflow communication: one workflow writes data, another reads it
• Simple lookups: store reference data (e.g. user preferences, config values) that workflows look up at runtime
• Audit logs: record every workflow action to a table for review or export

Creating a Data Table

1. Open Data Tables in the Workflows section of the sidebar
2. Click New Table
3. Define your columns: name, type (text, number, boolean, date, JSON), and whether the column is required
4. Save the table

Tables are workspace-wide and can be used by any workflow.

Reading Data

Use the Get Rows action (or Find First Row for a single result):

Input	Description
Table	Select the data table
Filters	Optional conditions, e.g. status = "pending"
Sort By	Column to sort results by
Sort Direction	Ascending or descending
Limit	Maximum number of rows to return

Get Rows outputs:

Output	Type	Description
rows	array	Matching rows as objects
rowCount	number	Number of rows returned

Find First Row outputs:

Output	Type	Description
row	object	The first matching row, or null if none found
found	boolean	Whether a matching row was found

Writing Data

Insert Row

Add a new row to the table.

Input	Description
Table	Select the data table
Row Data	Key-value pairs mapping column names to values

Outputs: rowId (the new row's ID), row (the full inserted row)

Upsert Row

Insert a new row, or update an existing one if a match is found. Useful for "create or update" patterns without needing a prior lookup.

Input	Description
Table	Select the data table
Match Columns	Which columns to use as the unique key
Row Data	Values to set (for both insert and update)

Update Rows

Update all rows matching a filter.

Input	Description
Table	Select the data table
Filters	Conditions identifying which rows to update
Updates	Column values to set on matched rows

Output: updatedCount, the number of rows modified

Delete Rows

Remove all rows matching a filter.

Input	Description
Table	Select the data table
Filters	Conditions identifying which rows to delete

Output: deletedCount, the number of rows removed

Count Rows

Count how many rows match a filter without returning the rows themselves. Useful for checking whether a table is empty or measuring progress.

Output: count, the number of matching rows

Viewing and Editing Data

Open any data table from Data Tables in the sidebar to view and manually edit its contents. You can:

• Browse, filter, and sort rows
• Manually add or delete rows
• Edit individual cell values inline
• Export the table to CSV

Best Practices

Use filters precisely. Get Rows without filters returns all rows. Add a Limit to avoid loading unexpectedly large datasets into a workflow node.

Use Upsert for idempotent writes. If a workflow might run multiple times for the same input (e.g. re-triggered by a webhook retry), use Upsert instead of Insert to avoid duplicate rows.

Clean up old data. Periodically delete processed rows with a scheduled workflow that deletes rows older than a threshold date, to keep tables from growing unboundedly.

JSON columns for flexible data. If the shape of a row's data varies (e.g. raw API payloads), use a JSON column rather than creating many optional text columns.