Pipedream Google BigQuery steps in n8n

BigQuery shows up in Pipedream workflows as the analytics sink: webhook events, Stripe payments or form submissions are streamed into a table, and scheduled queries feed reports. In Pipedream these steps live in the Google Cloud app. n8n has a dedicated Google BigQuery node with the two operations that matter.

The n8n side

Google BigQuery node (n8n-nodes-base.googleBigQuery):

• Execute Query: standard SQL, with options such as legacy SQL off, location, maximum results, dry run and query parameters.
• Insert: rows into a table, mapped from the incoming items (auto-map or chosen columns); options include skip invalid rows and ignore unknown values.

There is no BigQuery trigger node.

What maps where

Pipedream (app google_cloud)	In n8n	Status
Bigquery Insert Rows	Google BigQuery → Insert	manual
Run Query	Google BigQuery → Execute Query	manual
Create Scheduled Query	Schedule Trigger + Execute Query, or keep the BigQuery scheduled query	manual
List Dataset ID Options	not needed: the node lists projects, datasets and tables	—
BigQuery - New Row (polling)	Schedule Trigger + Execute Query (WHERE created_at > last run)	manual
BigQuery - Query Results (polling)	Schedule Trigger + Execute Query	manual
Upload An Object, Pub/Sub - Publish Message	Google Cloud Storage / Google Cloud Pub/Sub nodes or HTTP Request	manual

The converter has no BigQuery mapping yet. Each step arrives as a placeholder that keeps its props (project, dataset, table, rows or SQL) in a comment. The report lists a Google Cloud service account credential for the google_cloud app.

Google BigQuery credentials in n8n

• Service Account (recommended for automations): create a service account in Google Cloud, grant BigQuery Data Editor and BigQuery Job User on the project, download the JSON key, paste email and private key into n8n.
• OAuth2: acts as a user; fine for ad-hoc queries, less suited to unattended workflows.

Typical migration: Stripe payments to BigQuery

Pipedream: Stripe New Payment → Bigquery Insert Rows (analytics.payments). n8n: Stripe Trigger → Edit Fields (id, amount, currency, created as timestamp) → Google BigQuery → Insert (dataset analytics, table payments). Column names must match the table schema exactly.

Replacing BigQuery - New Row

1. Schedule Trigger every 15 minutes.
2. Execute Query: SELECT * FROM dataset.table WHERE inserted_at > TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 15 MINUTE).
3. Optionally Remove Duplicates on the row ID to cover overlapping windows.

Each poll is a billed query; partition or cluster the table on the timestamp column to keep scans small.

Google BigQuery migration pitfalls

• Streaming limits. insertAll takes at most 50,000 rows per request; Google recommends about 500 rows and around 10 MB per request. Batch large item lists before Insert.
• Streaming buffer. Rows inserted by streaming cannot be updated or deleted with DML for a while after insert.
• Types. Timestamps must be ISO strings or epoch seconds; nested RECORD fields need objects, not JSON strings.
• Location. Queries run in the dataset's location; set it in the node options for non-US datasets.

How to move your BigQuery workflows

1. Export from Pipedream.
2. Upload on /convert; the free report lists every BigQuery step with its SQL or rows.
3. Get the n8n JSON: 1 workflow is free; full export: $3 per workflow, $19 minimum — coming soon.

See the Pipedream to n8n guide. Related: PostgreSQL, Google Sheets, MySQL.