Retain Power Automate Run History Beyond 28 Days Using Metadata

Power Automate has become the backbone for many business processes, integrations, and automation scenarios across the Power Platform ecosystem. From approval processes to large-scale integrations, organisations increasingly rely on cloud flows to power mission-critical workloads.

Although Power Automate provides a comprehensive monitoring experience, one limitation quickly becomes apparent in production environments: cloud flow execution history is retained for only 28 days by default.

For many organisations, this introduces several operational challenges:

• Historical execution data is automatically removed after 28 days
• Troubleshooting failures that occurred months ago becomes difficult
• Long-term trend analysis is not possible
• SLA monitoring and compliance reporting become challenging
• Teams often resort to manual exports before execution history expires

Initially, organisations attempt to overcome these limitations through manual exports, Power BI snapshots, or custom logging implementations. While these approaches provide temporary visibility, they often increase maintenance overhead and introduce fragmented monitoring experiences.

To address this challenge, Microsoft introduced Cloud Flow Run Metadata in Dataverse, allowing organisations to persist cloud flow execution information beyond the default retention period.

In this article, we’ll explore how to extend the default flow run retention period using FlowRunTimeToLiveInSeconds, retrieve Cloud Flow Run Metadata from Dataverse, and build a lightweight dashboard for monitoring flow health and execution success rates.

Approach

Rather than relying solely on the Power Automate portal for short-term diagnostics, we can leverage Cloud Flow Run Metadata stored in Dataverse to build a historical monitoring solution.

Our approach consists of three steps:

1. Inspect the current retention period.
2. Update FlowRunTimeToLiveInSeconds to extend run history retention.
3. Retrieve flow execution metadata and surface it within a custom dashboard.

The dashboard presented in this article intentionally focuses on four key operational metrics

While simple by design, this dashboard serves as a foundation that organisations can later enhance with trend analysis, SLA reporting, execution duration insights, and failure analytics. Read More