The Chatbot analytics page in Applaud gives you clear insights into how employees are using your chatbots. It highlights which bots are most effective, which ones may need improvement, and how often they are used. By tracking user feedback, engagement, and sentiment, HR managers can identify where chatbots are delivering value and where they might be creating friction. These insights help you continuously improve employee support, streamline routine queries, and ensure your chatbots remain a trusted part of the digital workplace.
Metrics
The Chatbot analytics page provides key performance indicators to help you measure usage, effectiveness, and value. These metrics give you a quick overview of how chatbots are being used, the outcomes they generate, and the feedback received during the selected time period.
| Total users | Returning users | Conversations | Messages |
|---|---|---|---|
| The number of unique users sent or received a message during the selected period. | The number of unique users who engaged with a chatbot during the selected period but had not interacted with it in the previous six months. | The total number of conversations that had a message sent or received during the period. | The total number of messages sent or received during the period. |
| Hours Save | Cost saved | Sentiment | Feedback |
|---|---|---|---|
| The estimated time saved by chatbot interactions. Calculated using the number of messages with positive or neutral feedback (excluding negative feedback). For example, 100k messages equates to approximately $4m in savings. | This is the number of hours saved times $40, according to the calculation mentioned. | The average sentiment score of chatbot feedback comments submitted during the period. Scores range from -100% (very negative) to 100% (very positive). | This shows the number of positive and negative feedback given during the period. |
Best practices for using chatbot metrics
-
Track adoption trends
Monitor Total users and Returning users to understand how many employees are engaging with your chatbots. A rise in returning users often signals growing trust and usefulness. -
Spot workload reduction
Use Hours saved and Cost saved to demonstrate the tangible value of chatbots in reducing manual work, such as answering routine HR queries. This can help you build a stronger business case for expanding chatbot use. -
Identify improvement areas
Watch Sentiment and Feedback scores closely. Negative trends can highlight where employees find the chatbot unhelpful, giving you opportunities to refine responses or escalate to a human. -
Measure engagement depth
Compare Conversations and Messages to see whether users are having short, one-off exchanges or more prolonged interactions. More extended conversations may suggest higher engagement, but can also point to complexity in the process that could be streamlined. -
Review regularly
Run reports over consistent periods (monthly or quarterly) to track changes. This helps you identify whether new chatbot updates are improving the employee experience.
Most Effective and Least Effective
These blocks highlight how well your chatbot handles different topics, based on feedback sentiment. Topics are grouped using topic modeling techniques, rather than direct keywords, so the insights are generated automatically without prior setup.
By default, topics are sorted by their average sentiment score, which is calculated from feedback comments. Sentiment is not based on the number of “thumbs up” or “thumbs down” ratings alone but on the overall tone of the comments. The topics with the strongest positive or negative sentiment appear at the top of each list.
Each topic includes a sentiment bar:
- Far left (-100 to -16): Negative sentiment (for example, harmful or frustrating).
- Middle (-15 to 15): Neutral or balanced sentiment.
- Far right (16 to 100): Positive sentiment (for example, helpful or reassuring).
The Most Effective list shows topics where the chatbot received the most positive feedback.
The Least Effective list highlights topics where feedback was negative, such as comments flagged as harmful, untrue, or unhelpful.
You can adjust the sort order to view:
- Most Ratings (highest number of thumbs up/down)
- Most harmful
- Most untrue
- Most unhelpful
Selecting a topic row displays snippets of feedback comments related to that subject. These snippets provide context without showing full comments or personal information, which is automatically masked for privacy.
Example
An onboarding chatbot might appear under Most Effective if employees leave positive feedback about quick access to induction materials or IT setup instructions. However, if the same bot frequently struggles to answer payroll questions, those topics may appear under Least Effective, flagged as unhelpful or unclear.
Most Used Bots
This block shows which of your chatbots are used the most, giving you a quick view of their reach and impact. Bots are listed with key details, including:
- Bot name
- Type (Generative AI or Rules-based)
- Number of users who interacted with the bot
- Conversations handled during the period
- Likes and dislikes received as feedback
By default, bots are sorted by the number of users, but you can use the filter icon to reorder the list by other factors, such as conversations or sentiment.
Example
If you’ve deployed multiple chatbots, such as one for onboarding, one for payroll queries, and one for performance reviews, you can quickly see which bot is most used by employees. For instance, an Onboarding bot might top the list during hiring season, while a Payroll bot could show higher usage around payday.