Analytics & Reporting
Understand how your agents are performing, where customers are getting stuck, and how to improve resolution rates — all in one dashboard.
Key metrics
Conversation volume
Total conversations started over a time period, broken down by day, week, or month. Filter by agent or channel.
Average response time
Median time between a customer message and the agent's reply. Track this to catch latency regressions after model or knowledge base changes.
Resolution rate
Percentage of conversations resolved by the AI without human escalation. This is your primary automation health metric.
Sentiment trends
Positive, neutral, and negative sentiment over time. A rising negative sentiment rate often indicates a knowledge gap or a product issue.
Topic distribution
Automatically detected topics across all conversations (e.g. billing, returns, technical support). Use this to spot which topics drive the most volume.
Lead conversion funnel
Conversations that captured a lead, were scored, and converted — visible end-to-end for agents with lead capture enabled.
Accessing analytics
Go to Analytics in the sidebar
From your dashboard, click Analytics in the left sidebar. The overview dashboard loads with account-wide data.
Select a date range
Use the date picker to filter by last 7 days, 30 days, 90 days, or a custom range. All charts update instantly.
Filter by agent or channel
Use the agent dropdown to see metrics for a single agent. Filter by channel (widget, WhatsApp, Slack, etc.) to isolate performance.
Drill into conversations
Click any chart segment to see the underlying conversations. From there you can read transcripts and identify patterns.
Per-agent comparison
If you have multiple agents (e.g. one for support, one for sales), the comparison view shows them side-by-side so you can identify which is performing better.
Available per-agent metrics
Sentiment analysis
Every message is scored for sentiment in real time. The sentiment dashboard shows trends over time and surfaces conversations with strong negative signals so your team can review them.
Positive
Customer is satisfied or happy with the response
Neutral
Factual or informational exchange with no emotional signal
Negative
Customer is frustrated, confused, or dissatisfied
Tip: Set up sentiment-based escalation rules to automatically hand off conversations with negative sentiment to a human agent. Go to your agent settings → Escalation Rules → Add Rule → Sentiment is Negative.
Topic detection
SolveoAI automatically categorizes conversations into topics so you can see what customers are asking about most. Use this to prioritize knowledge base improvements.
Exporting data
Export analytics as CSV
Click the Export button on any analytics view to download the data as a CSV file. You can then import it into Excel, Google Sheets, or your BI tool (Tableau, Power BI, Looker).
Exportable data includes: conversation list, sentiment per conversation, topics per conversation, message counts, and response times.