Measuring contributor actions using log analytics

This article explains how to reliably measure contributor actions in Adobe Experience Manager (AEM) as a Cloud Service by combining access logs, audit logs, and other log sources with log analytics tools. It highlights the limitations of using access logs alone and provides recommendations for building accurate measurement models.

Description description

Environment

Adobe Experience Manager (AEM) as a Cloud Service

Issue/Symptoms

  • Organizations often use log analytics tools to track contributor actions such as page creation, modification, publication, and asset operations in AEM as a Cloud Service.
  • Relying solely on access logs can lead to inaccurate results because these logs capture HTTP requests, which may not correspond directly to completed business actions.
  • Publication events may be logged as requests to generic endpoints without specifying the exact content affected, making it difficult to identify which pages or assets were published.
  • Bulk or asynchronous operations may affect multiple items in a single request, and the logs may not provide clear mapping to business actions.

Cause

Access logs capture HTTP requests rather than confirmed business actions. As a result, contributor activities such as publication, asset operations, and bulk or asynchronous actions cannot always be accurately measured using access logs alone.

Resolution resolution

To accurately measure contributor actions in AEM as a Cloud Service, follow these steps:

  1. Do not rely solely on access logs for measuring contributor actions, as they primarily capture HTTP requests, not completed business operations.

  2. Combine access logs with additional sources:

    • Use audit logs to confirm completed actions and identify affected content paths.
    • Leverage workflow or event logs for operations involving asynchronous or bulk processing.
    • Where available, use AEM APIs to retrieve authoritative information about actions performed and their outcomes.
  3. Validate your measurement model through controlled testing:

    • For each contributor action (for example, page creation, modification, publication, unpublication, asset upload, deletion), perform a test and observe the resulting entries in access, audit, workflow, and event logs.

    • For each action, identify:

      • The user and timestamp
      • The action performed
      • The affected content path(s)
      • Whether the operation is synchronous or asynchronous
      • The indicator confirming success
      • The expected number of objects per request
      • How retries and duplicate events are represented
  4. Build a reference matrix mapping log entries to business actions. For example:

    • Page publication: Confirm with audit log entry showing action publish and the specific content path.
    • Asset deletion: Confirm with audit log entry for delete and affected asset path.
    • Bulk operations: Use workflow or event logs to identify all affected items.
  5. Update your log analytics queries based on the validated reference matrix to ensure KPIs are based on authoritative sources and accurately reflect completed business actions.

  6. Treat existing KPI results as indicative until validation is complete, especially for publication, bulk DAM operations, and asynchronous actions.

  7. Document the mapping between log entries and business actions for each scenario (Sites and Assets) to ensure ongoing accuracy.

  8. After updating your queries, verify the results by repeating controlled tests and confirming that the measured actions match the actual business operations performed.

recommendation-more-help
experience-cloud-kcs-help-kbarticles