This glossary explains the key statistics presented in Orlo's Chatbot Analytics Report, helping you understand your chatbot's performance.
Glossary 📖
Total Chatbot Sessions
This is the total number of individual interactions or conversations initiated with your chatbot during the selected time period. Each time a user starts a new chat with the bot, it counts as one session.
Chatbot Messages Handed Over vs. Not Handed Over
This metric shows the breakdown of chatbot interactions:
- Handed Over: The number of chat sessions that were transferred from the chatbot to a human agent. This often indicates the chatbot couldn't fully resolve the user's query.
- Not Handed Over: The number of chat sessions where the chatbot successfully addressed the user's query without requiring human agent intervention. This contributes to your chatbot's deflection rate.
Chatbot Feedback
This section reflects user feedback provided (if enabled) at the end of their interaction with the chatbot:
- Resolved: The number of chatbot sessions where the user indicated that their issue or question was successfully resolved by the bot.
- Unresolved: The number of chatbot sessions where the user indicated that their issue or question was not successfully resolved by the bot.
- No Feedback: The number of chatbot sessions where the user did not provide any feedback after interacting with the bot.
Top URLs Promoting Chatbot Messages
This lists the website URLs where the chatbot was most frequently accessed or initiated. This helps you understand which pages are driving the most chatbot interactions.
Topics Used
This provides an overview of the topics within your chatbot's Metabase that were triggered during user interactions:

Top 10 Topics
This section highlights the ten most frequently used topics by your chatbot during the reporting period, based on the number of times they were triggered. This provides a quick snapshot of the most common user inquiries.
Topic Feedback Score
This score reflects the average feedback rating for each specific topic used by the chatbot. It's calculated based on the "Resolved" and "Unresolved" feedback received for sessions where that topic was triggered. A higher score indicates that users found the chatbot's responses for that topic more helpful.