Meta Ads to BigQuery Pipeline

How to set up

Open the Select Accounts (Meta Ads) node and choose the account(s) you want to sync.
– You can select multiple accounts.
– All dimensions will include the account name and ID for easier identification.

Open the Create Dataset node, connect your BigQuery account, and select the GCP project where the dataset should be created.

Click Run to start the workflow.

Notes

Backfilling historical data
– To backfill, adjust the date range in the Select Date Range node.
– For large accounts, it’s recommended to backfill year by year or 6 months at a time to avoid hitting Meta Ads API limits.

Preventing duplicates
– The workflow includes a delete step that removes rows for the selected date range before inserting new ones.
– This means if the workflow fails today, you can safely rerun it — no duplicates will be created.
– You can also use this feature to refresh data for a specific range.

Dataset and tables
– The Create Dataset and Create Table nodes won’t cause errors on subsequent runs.
– If you want to save credits, you can delete unused datasets or tables, but make sure to update their references (dataset ID and table ID) in connected nodes.

Error handling

Continue on Fail is enabled in the Meta Ads node. If one account fails, the workflow continues with the remaining accounts.

Each request to the Meta Ads API retries up to 3 times before being marked as failed.

To log failed accounts, add a False path to the If Router node. Append the failed account details to a Google Sheet (or similar).

Attach a final node to the Done handle in the loop to email yourself a summary of failed accounts.

This template automates the full pipeline to move your Meta Ads performance data into BigQuery for reliable, up-to-date reporting. It creates a dedicated dataset, builds four partitioned tables (accounts, campaigns, adsets, and ads), ensures no duplicates by deleting existing rows for the selected date range, and then inserts fresh data.

How it works

You select a date range and the Meta Ads accounts you want to sync.
The workflow creates (or reuses) a BigQuery dataset.
It automatically creates the four required tables if they don’t already exist. All tables are partitioned by date to keep data organized and efficient.
For each table, existing rows in the selected date range are removed to prevent duplicates.
Fresh account, campaign, ad set, and ad-level performance data is inserted into BigQuery.

Who it’s for
This template is designed for marketers, analysts, and data teams who want a clean, automated pipeline to bring Meta Ads data into BigQuery for dashboards, cross-channel analysis, or advanced SQL queries — without the hassle of manual exports or duplicate data handling.

Requirements
To use this template, you’ll need to:

Connect your BigQuery account.
Connect your Meta Ads account(s).
Once connected, you can schedule the workflow to run daily so your BigQuery dataset always stays fresh.

Additional Notes

Backfilling data: To backfill, adjust the date range in the Select Date Range node. For large accounts, backfill year by year or 6 months at a time to avoid Meta Ads API limits.
Duplicate prevention: Each run automatically deletes rows in the selected date range before inserting new ones. This makes it safe to re-run failed workflows or refresh specific ranges without creating duplicates.
Dataset and table handling: The Create Dataset and Create Table steps won’t cause errors if they already exist. You can delete unused datasets/tables to save credits, but be sure to update references (dataset ID and table ID) in other nodes.
Error handling: If the workflow fails, you’ll automatically receive email alerts. No extra error monitoring is required.