Just noticed that a critical Data Action configured to update a custom attribute on the contact level is failing intermittently within our primary inbound sales flow. The specific Data Action is designed to capture the outcome of a third-party CRM verification step and write the result back to the contact object for subsequent routing decisions. The failure occurs approximately 15% of the time, resulting in a timeout error from the external endpoint. While the Architect flow logs indicate a partial failure where the data action returns a null response, the Performance dashboard metrics for ‘Data Action Success Rate’ do not seem to align with the observed failure rate in the flow logs. The dashboard shows a 98% success rate over the last 24 hours, which contradicts the manual review of failed interactions in the Conversation Detail view. The environment is configured in the EU-West region, and the Data Action is set with a 5-second timeout, which is the standard recommendation. However, the external API sometimes takes up to 4 seconds to respond, pushing the total execution time close to the limit. The question is whether the Performance dashboard aggregates the success metric based on the initial HTTP response code or the final data payload validation. If the dashboard only checks the HTTP 200 response, it might be missing the logical failures where the payload is empty or malformed. Additionally, there is no clear indicator in the Agent Performance view to highlight these partial failures, making it difficult for supervisors to identify agents impacted by these routing errors. The current workaround involves manual inspection of the flow logs, which is not scalable for our operation size. Is there a way to configure the Data Action to explicitly report a failure status to the dashboard metrics when the payload is invalid, rather than just relying on the HTTP status code? Or is there a specific filter in the Performance API that can isolate these partial failures based on the data action execution result? The lack of visibility into these edge cases is affecting our ability to optimize the flow logic and ensure accurate routing for high-value leads.