The Key Metrics
Your Analytics dashboard shows several metrics at a glance:| Metric | What It Measures | Why It Matters |
|---|---|---|
| Automation Rate | Percentage of tickets resolved without human intervention | Your primary success indicator |
| Total Tickets | All tickets processed by Cevro | Volume baseline |
| Automated Tickets | Tickets fully resolved by the AI | Direct measure of AI effectiveness |
| Escalated Tickets | Tickets handed off to human agents | Opportunities for improvement |
| AICSAT | AI Customer Satisfaction score (1-10) | Quality of automated resolutions |
| Resolution Time | Average time to resolve a ticket | Speed of service |
| Time Saved | Hours of agent work automated | ROI metric |
| Money Saved | Estimated cost savings from automation | ROI metric |
Automation Rate
The automation rate is the most important metric — it directly reflects how well Cevro is handling player support without human intervention:What’s a Good Automation Rate?
There’s no universal target — it depends on your setup maturity and the complexity of your support operations. However:- First week: Expect lower rates as you’re still building AIPs and identifying gaps
- After initial setup: 40-60% is typical for new deployments
- Mature deployments: 70-85% is achievable with well-tuned AIPs and comprehensive knowledge base
The automation rate improves over time as you add more AI Procedures, fill knowledge gaps, and refine instructions based on real conversations.
How to Improve Your Metrics
1. Review Unresolved Topics
The Unresolved Tickets tab in Analytics shows exactly what players are asking about that Cevro can’t handle yet. Each topic is an opportunity:- High trigger count? → Create a new AIP or add a knowledge base article
- Related to an existing AIP? → Your instructions may need updating
2. Review Escalated Conversations
Check the Conversation Logs for escalated tickets. Common fixable causes:| Escalation Reason | Fix |
|---|---|
| Knowledge gap | Add a KB article covering the topic |
| Missing AIP | Create an AIP for this ticket type |
| Incomplete instructions | Update the AIP’s SOPs with the missing scenario |
| Tool error | Check tool configuration and back-office connectivity |
3. Monitor AICSAT Scores
Low AICSAT scores on specific AIPs indicate quality issues. Review those conversations to understand whether the AI’s responses are accurate, helpful, and on-tone.4. Use the Playground
Before deploying changes to live traffic, test them in the Playground. This lets you simulate player conversations and verify that your updates work as expected.Filtering and Drill-Down
Use the filters at the top of the Analytics page to narrow your view:- Date range — Compare performance across time periods
- Brand — See how individual brands perform
- Category — Focus on specific AIP categories
Continuous Improvement Cycle
Prioritize by impact
Focus on the highest-volume unresolved topics first — they’ll have the biggest effect on your automation rate.
Related
- Understanding Metrics — Detailed breakdown of every metric and column
- Topic Monitoring — Working with auto-generated topics
- Conversation Logs — Reviewing individual conversations
- Delegation vs Escalation — Understanding why tickets are handed off