This page will walk you through the different components and sections of our Analytics page, which you can use to analyse and optimise your bot in the future.
The dashboard is the main page in our Analytics. It shows a lot of important information about your bot in a quick and simple overview:
Let's go through all the components one by one, and see what they exactly mean for your bot. Each section also includes detailed information about how that specific component can be interpreted and used to improve your bot.
With these filters you can adjust the information shown on the dashboard, and create specific views for your bot. This provides valuable information to see how your bot progresses over time.
'All Channels' allows you to filter on specific channels, and 'All languages' shows information about your bot in different languages.
The timestamp of the data you see in the analytics dashboard is based on the timezone of your browser.
The 'User Messages' section shows an overview of how many messages are:
- Understood: the user expression is recognized correctly and the corresponding intent was triggered
The first 'Conversations' component shows the distribution of bot conversations in all available languages.
Switching from 'Languages' to 'Channels' will show the use of the bot over different channels (if these are used of course).
The second 'Conversations' component shows the human offloading, average messages per conversation, and the average length of conversations:
- Human agent takes over: a conversation is marked as a human handover when it includes a triggered "Send to offload provider" action
- Messages per conversation: this category represents the most occurring length of a conversation in number of messages for interactions between your bot and its users.
- Conversation duration: this category represents the most occurring length of a conversation in number of minutes for interactions between your bot and its users
'Most triggered intents' shows the top 5 intents that were triggered most often, including the number of calls and their percentages.
'Active customers' shows when your users are most active and interact with the bot.
'Flow - Most popular bot dialogs' shows which route most users take in your bot. In the example above, 15.9% of users first go to the website router, and the same amount of users first go to the introduction or to the 'Intro Hello' intent. 3.5% immediately goes to offloading and 3.5% got a 'not understood' message.
Hover over one of the specific bot dialogs to show the option 'Open'. Now click again to see a detailed overview of the exact flow the user is following: