Exported asset metadata reports include excessive predictive tag columns in AEM Assets

In Adobe Experience Manager as a Cloud Service (AEMaaCS), exporting asset metadata generates reports with an unexpectedly large number of columns due to predictive tags created by Adobe Sensei. These excessive columns exceed spreadsheet limits and complicate metadata mapping for integrations. To fix this, exclude predictive tag fields from metadata exports.

Description description

Environment

  • Product: Adobe Experience Manager as a Cloud Service (AEMaaCS) – Assets
  • Constraints: Predictive tags enabled by default after migration to AEM as a Cloud Service

Issue/Symptoms

  • Exported asset metadata reports include thousands of columns generated from individual predictive tags.
  • The number of columns exceeds spreadsheet application limits, such as Excel’s 16,384 column maximum.
  • Exports become unstable or incomplete because column limits are exceeded.
  • Metadata mapping for integrations becomes difficult to manage.

Resolution resolution

To exclude predictive tag fields from exported asset metadata reports, follow these steps:

  1. Access the OSGi configuration for your AEM as a Cloud Service environment.
  2. Locate the configuration com.day.cq.dam.core.impl.metadata.AssetMetadataExporterNew.
  3. Find the property excludedPropertyNamePatterns.
  4. Add the value predictedTags/.* to excludedPropertyNamePatterns to exclude all predictive tag fields from future exports.
  5. Apply and save the configuration changes.
  6. Test the export process in a lower environment to confirm that predictive tag fields are excluded.
  7. Promote the configuration change to production after successful testing.
  8. Verify that new exports no longer include excessive predictive tag columns.

Notes:

  • This change only affects which fields appear in exported metadata reports.
  • Predictive tags continue to be stored in AEM Assets after applying this configuration.
  • AEM Assets does not provide a configuration option to change how predictive tags are stored.
recommendation-more-help
3d58f420-19b5-47a0-a122-5c9dab55ec7f