Can anyone clarify the reconciliation logic between the execution time reported in the Architect Flow Performance dashboard and the actual latency observed when invoking a Data Action via the Engage API? The organization operates in EU-West BYOC. During recent load testing, the Performance dashboard indicates an average execution time of 250ms for a specific HTTP-based Data Action. However, tracing the request via the Engage API reveals a total round-trip time of 1.2s. The Data Action is configured with a timeout of 5s, which is not being triggered. The discrepancy suggests additional overhead between the flow execution engine and the external service call, or potentially a misalignment in how ‘execution time’ is defined within the performance metrics. The flow utilizes the latest version of the Data Action feature, and no errors are logged in the conversation detail view. The environment is standard BYOC with no custom network restrictions affecting outbound HTTP traffic. Understanding this variance is critical for accurate SLA reporting and capacity planning. Is there a documented breakdown of the components contributing to the total latency, or is the dashboard metric intended solely for internal processing time?