Google BigQuery - Select rows using custom SQL action (batch)
The Select rows using custom SQL batch action allows you to run a custom SQL query and retrieve matching rows from Google BigQuery. This action supports parameterized queries using bind variables. If the query runs longer than 60 seconds and Wait for query to complete? is set to No, the action returns the job ID and completes asynchronously. Use the Get batch of rows by Job ID action to retrieve the results later.
50,000 RECORD LIMIT
Starting February 8, 2026, the Select rows using custom SQL action returns a maximum of 50,000 records for all new recipes and existing recipes that don't currently exceed this limit. Starting March 15, 2026, this limit extends to all recipes using this action, including existing recipes that currently exceed the limit.
Recipes configured to retrieve more than 50,000 records will be truncated to this limit.
This change improves platform reliability and performance while preventing large queries from overloading database connections or impacting upstream systems.
Input
| Input field | Description |
|---|---|
| Project | Select the project to bill for the query. |
| Query | Enter the SQL query to run. Use the @ symbol to declare bind variables. |
| Parameters | Map bind variables used in the query. For example, for select * from table where id = @id, provide a parameter named id. |
| Output fields | Define the expected output fields. The names must match the columns returned by the query. |
| Location | Select the region to run the query. Optional in most cases. |
| Wait for query to complete? | Set to true to wait for the full query execution. Set to false to return only the Job ID after 60 seconds. Use the Get batch of rows by Job ID action to retrieve the results. |
Output
| Output field | Description |
|---|---|
| Output contains rows? | Indicates whether the result includes row data. |
| Rows | An array of row objects. Each datapill corresponds to a column in the result. |
| Kind | The type of job executed. This should be queryResponse. |
| Job reference | Includes the project ID, job ID, and location of the job. Use this with the Get batch of rows by Job ID action to retrieve more rows. |
| Total rows | Total number of rows returned by the query. |
| Page token | Token to retrieve the next page of results. |
| Schema | The schema of the query result, including column names and types. |
| Etag | A hash representing the result schema. |
| Total bytes processed | The number of bytes processed by the query. This counts toward your BigQuery billing. |
| Job complete | Status of the query job. Returns Done if the job completed successfully. Returns Pending if the job is still running. |
| Cache hit | Returns true if the query result was served from cache. Returns false otherwise. |
Last updated: