Production monitoring · Autessa Evaluations
Quality dashboards
Quality dashboards show how your agents are doing over time: eval pass rates by day, eval scores over time, conversations per day by channel, how human verdicts compare with automated ones, and what your response checks and the AI Firewall caught. They work for one agent or all of them.
The problem
Leaders ask “is it getting better?”
A pile of individual evaluations doesn’t answer that. Trend charts do, and they need to cover test runs and live conversations together, so improvements and slips are visible to people who will never open a transcript.
What you get
- Eval pass rates by day: the share of test cases and graded conversations that passed.
- Eval scores over time, per dimension.
- Conversations per day, by channel.
- Human vs automated: the outcome mix from reviewers and from the automated judge, side by side.
- What your checks caught: replies response checks caught each day, by what happened to the reply, and messages the AI Firewall caught.
- Every chart has a Show the numbers table view.
- The workspace dashboard shows a color-coded success rate for each active agent over three months, with a link into its history.
How it works
How does quality dashboards work in Autessa Evaluations?
Choose an agent or all agents
Pick the scope and the date range.
Read the trends
Pass rates, scores and conversation volume, by day.
Show the numbers
Switch any chart to a table for exact values.
Drill in
Jump into Agent History for the conversations behind a dip.
Every option
Charts and cards
| Option | What it does |
|---|---|
| Eval pass rates by day | Test cases and graded conversations that passed, per day. |
| Eval scores over time | Dimension scores over time. |
| Conversations per day, by channel | Conversation volume by channel. |
| Human vs automated | Outcome distributions from human and automated evaluations, each with its own count. |
| Status distribution | Successful, with caveats, unsuccessful and not evaluated. |
| What your checks caught | Response check outcomes per day, and AI Firewall catches. |
| Success rate bands | 80% and up: excellent; 60–79%: good; under 60%: needs improvement. |
Permissions
Who can do what?
| Who | Can |
|---|---|
| Anyone who can read an agent | See its dashboards; the workspace view covers agents active in your company. |
Connected
How it connects to the rest of Evaluations
FAQ
Questions about quality dashboards
Do the dashboards include test runs?
Yes. Eval pass rates by day combine test cases and graded conversations.
Can I see exact numbers, not just charts?
Yes. Every chart has a Show the numbers table view.
What counts as a good success rate?
The product bands it as 80% and up excellent, 60–79% good, and under 60% needs improvement.
100% customizable, code and all
Make Autessa Evaluations fully your own.
Use it out of the box, then change anything. Ask Autobot, Autessa’s AI builder, for a new field, screen, report, workflow or agent and it builds it into your copy of the app. Or open the code and edit it yourself. Nothing is locked: the app, its data and its code are yours.
Start free and make it yours