Agentic Evaluation
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
3 hours ago
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
3 hours ago
I've run into this exact issue before. The overall score not showing up after automated evaluation usually comes down to a few things.
First thing to check is your evaluation criteria configuration. If any of the criteria are missing required fields or have scoring conflicts, the overall calculation just won't complete. Go back into your agent's evaluation setup and make sure each criterion has proper weight assignments and scoring ranges defined.
The other common culprit is data access. The evaluation process needs to read through conversation records and sometimes hits permission issues or scope problems. Check if your agent is running in the right application scope and that the evaluation user has proper access to the conversation data.
Without seeing the specific error details in your screenshots, I'd also suggest looking at the system logs around the time the evaluation ran. Navigate to System Logs > System Log > All and filter by the timeframe when you triggered the evaluation. You'll likely see more detailed error messages there that point to the root cause.
One quick test - try running a manual evaluation on a single conversation first to see if that works. If manual works but automated doesn't, it's probably a scheduling or batch processing issue.
What type of agent are you evaluating, and are you seeing any specific error messages in the logs? That would help narrow it down further.

