Google Ads connector concepts

Beta

This feature is in Beta. Workspace admins can control access to this feature from the Previews page. See Manage Databricks previews.

Learn how the managed Google Ads connector in Lakeflow Connect works.

How the connector works

The Google Ads connector ingests data from Google Ads accounts into Databricks using the Google Ads API. The connector supports two types of tables:

• Resource tables: These tables contain configuration data about your Google Ads account, like campaigns, ad groups, and ads. Resource tables do not support incremental ingestion; they are fully refreshed on each pipeline run. They also do not support history tracking (SCD type 2).
• Report tables: These tables contain performance metrics, like impressions, clicks, and conversions. Report tables support incremental ingestion with a configurable lookback window to capture late events and attribution updates.

For a list of supported tables, see Google Ads connector reference.

Account hierarchy

The Google Ads connector works with the following account structure:

• Manager Account ID (MCC): A 10-digit identifier for a Google Ads Manager account that manages one or more Google Ads accounts. One pipeline maps to at most one Manager Account ID.
• Customer Account ID: A 10-digit identifier for a specific Google Ads account. If your Manager Account ID maps to multiple Customer Account IDs, you can ingest from those different Customer Account IDs within the same pipeline.

Google Ads organizes advertising data in the following hierarchy:

• Manager account: Umbrella account to manage multiple client ad accounts, users, and billing.
  • Ad account: Individual advertising account.
    • Campaign: Top-level organizational unit for ads.
      • Ad Group: Collection of ads that share targeting settings.
        • Ad: Individual advertisement.
          • Asset: Reusable creative components (headlines, images, logos, sitelinks) that can be attached to ads or campaigns.

Data ingestion patterns

Historical data sync

By default, the connector ingests two years of historical data for report tables. You can configure this using the sync_start_date parameter. There is no maximum limit.

Incremental updates and lookback windows

Google Ads conversion data can change for up to 90 days after the fact. For example, a user might convert days after clicking an ad, or Google Ads might recalculate attribution. To capture these changes, the connector re-checks the last 30 days of data by default during each pipeline update.

You can customize this lookback window using the lookback_window_days parameter. Consider your organization's conversion attribution window when setting this value. For example, if your organization uses a conversion window of 60 days, set lookback_window_days=60 to ensure that late conversions and attribution updates are reflected in your data.

Money storage in Google Ads

Google Ads stores monetary values in micros to avoid rounding issues. One million micros equals one unit of currency. For example:

• 1,000,000 micros = $1.00 USD
• 5,500,000 micros = $5.50 USD

When working with cost and budget fields in your destination tables, divide by 1,000,000 to convert from micros to your currency.