Skip to main content

​
Requirements

  • Google Cloud Platform (GCP) account
  • GCP project with BigQuery resource enabled
  • Chosen region for the dataset (cloud storage bucket) and BigQuery to reside
  • A user with admin access to the project’s BigQuery resource

​
Step 1: Create a BigQuery Dataset

Create a BigQuery dataset in your target GCP project. We recommend colocating within a single region, but support multi-region setups.
⚠️ Note on regions: Google recommends colocating your BigQuery dataset within a single region for better performance. See Google documentation for details.

​
Step 2: Connect Rockerbox to BigQuery

  1. Open the Warehousing setup page Rockerbox and choose BigQuery → Connect to BigQuery.
  2. Enter your dataset region, project ID, and dataset ID, then click SetupBigQuery.

​
Step 3: Copy the Rockerbox Service Account

After initialization, Rockerbox displays a unique service account email. Click the icon next to it to copy the address.

​
Step 4: Grant Dataset Permissions in GCP

In GCP, grant that service account with the DataEditor role roles/bigquery.dataEditor to your BigQuery dataset so Rockerbox can create tables.

​
Step 5: Grant Team Access (Optional)

Back in the Rockerbox UI, add users, groups, and/or service accounts from your organization that should access the shared data in BigQuery.

​
Step 6: Notify Rockerbox (One-Time Enablement)

Tell your Rockerbox support rep that you’ve completed the steps above. Due to a BigQuery limitation, Rockerbox must sync underlying datasets first before you can create tables in the Rockerbox UI. Rockerbox support will confirm when you can proceed to the next step.

​
Step 7: Sync Rockerbox Data Sets

  • Choose which datasets to share:
    • Platform Data: Select the ad platforms you actively use.
    • Rockerbox Data: Select Conversion and any Log Level MTA datasets for each conversion event that you need.
  • Click Sync this dataset when ready.
📦 aggregate_mta and taxonomy_lookup tables are automatically created in the data share.

​
⚠️ Backfill Notes

  • Platform Performance Schemas: No data is automatically backfilled. This can be backfilled on a limited basis upon request to support@rockerbox.com.
  • Rockerbox First Party Data Schemas: For each conversion dataset, Rockerbox will backfill based on the conversion event “First Reporting Date” in Rockerbox
    💡 If no date is set, Rockerbox backfills one day of data. This process can take up to 24 hours.

​
Step 8: Query your new data tables

​
Sample queries: aggregate_mta table

Test setup 
SELECT * FROM `<project>.<dataset>.aggregate_mta` LIMIT 1;
Report even weight attributed conversions and spend by conversion event + tier_1 (most broad reporting taxonomy dimension) 
SELECT 
  conversion_event_name,
  tier_1,
  SUM(even) AS conversions_even,
  SUM(included_spend) AS spend
FROM `<project>.<dataset>.aggregate_mta`
WHERE 
  date >= DATE_SUB(CURRENT_DATE(), INTERVAL 7 DAY)
  AND date < CURRENT_DATE();

​
BigQuery Table Limitations

For MTA and Conversions schemas in BigQuery, nested columns—url_parameters, utm_parameters, and additional_attributes—won’t auto-expand when new parameters appear. Contact Rockerbox if you need newly added parameters reflected in those fields.