Quick Audience:User Insight FAQ

How can I use data imported by others for analysis?

All members except the administrator and data importer must be authorized before they can use imported data.

Authorization is granted at the tag, AIPL model, or RFM model level — not by table. Tags come from the imported user tag table, which includes custom tags and user attributes derived from the imported data. AIPL models and RFM models are built from the user behavior table, order details table, or order summary table.

There are three types of permissions:

• Use permission: the permission to use tags in scenarios such as perspective analysis, audience filtering, Kafka push, and marketing.

• Management permission: the permission to edit, authorize, and use the authorized tag.

• Row-level security: grants permissions on specified rows to enforce data isolation.

To grant permissions, you must be an administrator, a data importer, or a member with management permission. For details, see Authorize.

How do I synchronize the latest data in a computing source?

After you create an AIPL model, RFM model, or audience, updates to the underlying computing source are not automatically reflected in Quick Audience. To bring in the latest data, re-import the data table and then update the relevant models, custom tags, and audiences. Without this, your analyses will be based on stale data.

Here is what you need to update:

• Re-import the data table:

  • Schedule imports for user tag tables, user behavior tables, statistics tables, order details tables, and order summary tables using daily scheduling, hourly scheduling, or API-triggered scheduling. For details, see Scheduling Tasks.

• Update the model:

  • Manually update RFM models and AIPL models, or configure them to update based on the underlying data schedule. For details, see Manage AIPL Model and Manage RFM Model.

• Update custom tags:

  • Manually update custom tags, or set them to update on a daily, weekly, or monthly basis. For details, see Update Custom Tags.

• Update audience:

  • Manually update an audience, schedule updates based on underlying data, or set weekly or monthly updates. For details, see Audience Updates and Snapshots.

  Note

  Updating an RFM model, AIPL model, custom tag, or audience does not automatically trigger the source data table import. Schedule the data table import first, then run the update after the import completes. This ensures the latest data from the computing source is reflected in your models and audiences.

Recommended setup order:

1. Start by configuring the scheduled import for the underlying data table:

   • If the table is updated on a fixed schedule, use periodic scheduling.

   • If the table is updated irregularly, use manual scheduling or API-triggered scheduling — call the API after the underlying processing task completes.

2. Set your audiences, RFM models, and AIPL models to follow the underlying data schedule. Each time the data table imports successfully, these objects update automatically to reflect the new data.

What causes failures in data table import, AIPL models, RFM models, custom tags, and audience updates?

When a data table import or an AIPL model, RFM model, custom tag, or audience update fails, a failure icon appears. Hover over the icon to see the specific error.

Common causes of failures:

• Missing data permissions: the account authorization settings of the computing source may have changed. Sign in to the computing source as an administrator and verify the account authorization.

• Underlying data table not found: the data table may have been renamed or deleted. Check whether the underlying data table still exists under the expected name.

• Computing source password changed: the new password was not updated in the Quick Audience computing source configuration, causing all dependent data tables and audiences to fail. Ask the administrator to update the computing source password in Quick Audience.

• SQL execution failed: resources may be temporarily occupied. Wait and try again later.

What are the differences between ID encryption options when importing a data table and when pushing to a data bank or Damengpan?

The two stages — import and push — have different encryption options and behaviors.

When importing a data table to Quick Audience, set the ID encryption method to Unencrypted, MD5, SHA256, or AES. This tells Quick Audience how the raw data is already encrypted before import.

When pushing an ID to a data bank or Damengpan, select MD5 or SHA256 to specify the encryption method for the outbound ID. Quick Audience handles the conversion automatically based on the import-time encryption:

• If the raw ID was imported as unencrypted (original text), Quick Audience encrypts it using the selected MD5 or SHA256 method during the push.

• If the raw ID was imported with AES encryption, Quick Audience decrypts it first, then re-encrypts it using the selected MD5 or SHA256 method during the push.

• If the raw ID was imported with MD5 encryption, only MD5 is available as the push encryption method. Quick Audience keeps the existing MD5 encryption unchanged.

• If the raw ID was imported with SHA256 encryption, only SHA256 is available as the push encryption method. Quick Audience keeps the existing SHA256 encryption unchanged.