Analytics
Track conversation volume, AI performance, response times, lead conversion, and channel health — all in one dashboard.
Understanding your performance
The Analytics page (at Insights > Analytics) gives you a data-driven view of how your AI assistants and human team are performing. Use it to spot trends, identify problems, and make informed decisions about your configuration.
The dashboard is organized into several sections, each focused on a different aspect of your operation.
Conversation overview
The top section shows your high-level conversation metrics for the selected time period:
| Metric | What it tells you |
|---|---|
| Total conversations | How many conversations were started across all channels |
| AI-handled | How many were fully resolved by the AI without human intervention |
| Escalated to human | How many were handed off to a human agent |
| Resolved | How many were marked as resolved (by AI or human) |
| Average response time | How quickly conversations received their first response |
| Average resolution time | How long it took to resolve a conversation from start to finish |
Use the time range selector at the top to filter by day, week, month, or a custom date range. Use the channel filter to focus on a specific channel (chat, SMS, email, voice, or social).
AI performance
This section focuses specifically on how your AI assistants are performing:
- Handoff rate — The percentage of conversations the AI escalated to a human. A consistently high handoff rate may indicate gaps in your training context or playbooks.
- Playbook completion rate — How often the AI successfully completed its assigned playbook goal. Low completion rates often point to unclear instructions or customers who are going off-script.
- FAQ hit rate — How often the AI found a relevant answer in your training context. A low hit rate means customers are asking questions your knowledge base doesn't cover.
The FAQ miss report is one of the most actionable parts of Analytics. It shows you the questions your AI couldn't answer — which is exactly what you should add to your Training Context next.
Lead qualification
If you're using Autoch.at for lead generation, this section shows you how your qualification funnel is performing:
- Leads started — Conversations where the AI began a qualification playbook
- Leads completed — Conversations where the customer provided all required qualification data
- Qualified leads — Conversations where the customer met your qualification criteria
- Conversion rate — Qualified leads as a percentage of conversations started
Channel breakdown
See how each channel is performing individually. This helps you understand where your customers prefer to communicate and where your AI is most effective.
Team performance
If your workspace has human agents, this section shows individual and team-level metrics:
- Conversations handled per agent
- Average response time per agent
- Resolution rate per agent
This data is useful for identifying top performers and spotting agents who might need additional support or training.
Exporting data
Click the Export button to download any report as a CSV file. This is useful for sharing with stakeholders or importing into your own analytics tools.
Last updated 3 weeks ago
Built with Documentation.AI