home

Applications

ReltioProductsDeveloper resourcesRolesApplicationsObjectives
ReltioProductsDeveloper resourcesRolesApplicationsObjectives

Unify and manage your data

Show Contents

thumbnailApplications

Zero-copy integration from Google BigQuery overview

Configure zero-copy integration from Google BigQuery using the Console

Learn how to configure zero-copy integration from Google BigQuery using the Console.

Use Data Sharing in the Console to configure the BigQuery connection and interaction mappings for zero-copy integration from Google BigQuery. Reltio uses the connection and mappings to query interaction data in BigQuery. Matching rows appear in the interaction timeline of related profiles, and Reltio does not copy the data. You can also enable zero-copy segmentation on the same connection.

The following roles can configure zero-copy integration from Google BigQuery using the Console:

Prerequisites

Before you begin, confirm the following requirements:

  • Your organization must have a data integration add-on. Data Sharing appears in the Console only with this add-on. For more information, see Reltio data integrations—add-ons.

  • Your Reltio tenant must have zero-copy interaction integration enabled.

  • You must have ROLE_ADMIN_TENANT or an equivalent custom role with permission to manage data integrations.

  • Your Google Cloud project must have BigQuery enabled.

  • You must know the BigQuery project ID, dataset name, and the table or view that contains the interaction data.

  • You must have a Google Cloud service account created and managed by your organization. Reltio impersonates this service account and does not use a service account key.

  • You must have permission to manage access to the Google Cloud project and the service account.

  • The interaction type that you want to map must exist in your tenant.

  • The BigQuery table or view must contain crosswalk values that also exist on related mastered profiles in Reltio.

  • To enable zero-copy segmentation, your tenant must have a Data Pipeline Hub adapter of type gbq configured. Its export dataset must be in the same location as the BigQuery connection.

To configure zero-copy integration from Google BigQuery using the Console
  1. In the Console, under Tenant Management, select Data Sharing, and then select BigQuery.
    The BigQuery page shows the connection status at the top right. If the tenant has no connection, the status is Not connected.
  2. Open the BigQuery connection configuration.

    • If the tenant has no connection, select CONNECT.

    • If a connection exists, select the settings icon next to Connected.

    The Authentication configuration - BigQuery dialog opens. The Access section shows the Reltio service account and the Google Cloud roles that the integration requires.
  3. In Access, select the copy icon next to the Reltio service account, and then grant the following access in Google Cloud.
    1. On your Google Cloud service account, add the Reltio service account as a principal and grant it the Service Account Token Creator role (roles/iam.serviceAccountTokenCreator).

      Use a separate service account for each tenant.

    2. On your Google Cloud project, grant your service account the BigQuery Job User role (roles/bigquery.jobUser).
    3. On the dataset, or on each mapped table or view, grant your service account the BigQuery Data Viewer role (roles/bigquery.dataViewer).
    4. To use zero-copy segmentation, on the dataset that the Data Pipeline Hub adapter exports to, grant your service account the BigQuery Data Editor role (roles/bigquery.dataEditor).

      Reltio reads the entities view of the Data Pipeline Hub export and creates a segment_results table in this dataset.

  4. In the Authentication configuration - BigQuery dialog, in BigQuery project details, enter the project details.

    • GCP project ID: The Google Cloud project that holds the dataset that Reltio reads from. The ID must be 6–30 characters, start with a lowercase letter, contain only lowercase letters, digits, or hyphens, and not end with a hyphen.

    • Dataset: The BigQuery dataset that contains the interaction tables or views.

    • Location: Optional. The BigQuery dataset location. Leave the field empty to use the dataset default.

  5. In Impersonation, enter the impersonation details.

    • Customer service account: The email address of your Google Cloud service account that Reltio impersonates to run queries.

    • Query timeout (ms): The maximum time, in milliseconds, that Reltio waits for a BigQuery query to complete. The default is 3000.

  6. Optional: In Zero-copy segmentation, turn on Enable zero-copy segmentation, and then select a Data Pipeline Hub adapter.

    Turn on this option to read segment results from a Data Pipeline Hub BigQuery export, in addition to interactions. The option is off by default. When the option is off, Reltio uses the connection for zero-copy interactions only.

    The Data Pipeline Hub adapter list shows the options of type gbq configured for your tenant. If your tenant has none, the dialog shows a message instead of the list. Configure a Data Pipeline Hub adapter of type gbq, and then return to this step.

    To stop using the connection for zero-copy segmentation later, open the connection configuration, turn off Enable zero-copy segmentation, and then select CONNECT. Reltio removes the segmentation setting and the selected Data Pipeline Hub adapter from the connection.

  7. Select CONNECT.
    Reltio saves the connection. When the connection is working, the status at the top right of the BigQuery page is Connected. If the connection fails, Console shows the error message returned by Reltio. Correct the connection details or the Google Cloud access, and then select CONNECT again.
  8. On the BigQuery page, select + CREATE NEW.
  9. In the Create new dialog, select the Reltio interaction type and the BigQuery table that contains the interaction data.

    In the BigQuery table list, expand the project and the dataset, and then select the table or view. You can also search for the table by name.

  10. On the ATTRIBUTES tab, select the source columns for Interaction timestamp and Row key.

    • Interaction timestamp: The column that determines the order of interactions in the timeline.

    • Row key: A unique, non-null column that identifies a row. The row key keeps paging exact when timestamps repeat.

  11. On the ATTRIBUTES tab, select a source column for each interaction attribute that you want to return.
  12. On the MEMBER ENTITIES tab, select the source column that contains the crosswalk ID for at least one member type.
  13. Select SAVE.
    The mapping appears in the Interaction mappings list.

Result

Reltio saves the BigQuery connection and the interaction mapping. The Interaction mappings list shows the mapping with its Interaction type, Source table, and Last updated values. You can edit or delete a mapping by using the edit and delete icons on its row. To save changes to an existing mapping that has no row key, select a source column for Row key first.

When interactions are requested for a related profile, Reltio queries the mapped BigQuery table or view. Matching rows appear in the interaction timeline, and Reltio does not store the BigQuery data.

Verification steps

To confirm that the integration returns interaction data, perform the following steps:

  1. On the BigQuery page, confirm that the status at the top right is Connected.

  2. Confirm that the mapping appears in the Interaction mappings list.

  3. Open a profile whose crosswalk ID exists in the mapped BigQuery table or view, and confirm that the matching rows appear in the interaction timeline.

Updated on October 30, 2026

Footer

You've found your way to a temporary environment. Content is not complete or final. Looking for Reltio documentation? Go to docs.reltio.com.

© 2026 The RELTIO® mark is a registered trademark of Reltio, Inc. in the United States and other countries; DataDriven, Reltio Context Intelligence Platform and Reltio Data Cloud are trademarks of Reltio, Inc. Other names or marks are trademarks of their respective owners.

Terms and Conditions, Privacy Policy, and