Help Desk Guide for Messaging Logs
View and analyze conversation history and message performance
The Messaging Logs feature provides a comprehensive view of all conversations and messages sent through your AI assistants. With real-time analytics, powerful filtering, and sentiment analysis, you can monitor performance, troubleshoot issues, and gain valuable insights into your messaging operations.
Message Analytics Dashboard
At the top of the Messaging Logs page, you'll find the Message Analytics section displaying real-time statistics across multiple key metrics:
Assistant Selection
Use the Assistant dropdown to view logs for a specific assistant or select "All Assistants" to see aggregate data across your entire account.
Analytics Metrics
The dashboard displays the following real-time statistics:
Total Conversations
The cumulative number of unique conversation threads. Each conversation represents a distinct interaction with a user, helping you understand your total engagement volume.
Total Messages
The total count of individual messages sent and received across all conversations. This metric gives you insight into messaging activity and volume.
Unique Users
The number of distinct users who have engaged with your assistant. This helps you track your reach and audience growth over time.
Avg Response Time
The average time it takes for your assistant to respond to user messages. Measured in seconds, this metric is crucial for monitoring performance and user experience.
Today's Chats
The number of conversations that occurred today. This provides a quick snapshot of current-day activity and engagement levels.
Active (24h)
The number of active conversations in the last 24 hours. This metric helps you monitor recent engagement trends and peak activity periods.
Negative Sentiment
The count of conversations where negative sentiment was detected. This alerts you to potentially dissatisfied users who may need immediate attention or follow-up.
No Answer Found
The number of times your assistant couldn't provide an answer. This metric is critical for identifying knowledge gaps in your assistant's training that need to be addressed.
Positive
The count of conversations with positive sentiment detected. This indicates satisfied users and successful interactions with your assistant.
Successfully Answered
The number of queries where your assistant successfully provided a helpful answer. This is a key performance indicator for measuring assistant effectiveness.
Searching Conversations
Below the analytics dashboard, you'll find a powerful search bar that allows you to quickly locate specific conversations:
Search conversations by message content...
How to use: Enter keywords, phrases, or specific terms from messages you're looking for.
Purpose: This search function scans through message content to help you quickly find relevant conversations without manually browsing through logs. Perfect for investigating specific user queries, following up on issues, or auditing particular interactions.
Filtering Messages
The Messaging Logs page includes powerful filtering options to help you narrow down and analyze specific subsets of your data:
Date Range
Description: Filter conversations by specific date ranges using the calendar inputs.
Quick Filters: Use the convenient preset buttons for instant filtering:
- Today - View only today's conversations
- 7 days - Show conversations from the past week
- 30 days - Display the last month of activity
Purpose: Analyze messaging patterns over specific time periods, investigate issues that occurred on particular dates, or generate reports for defined timeframes.
Sentiment
Description: Filter conversations by detected sentiment (All sentiments, Positive, Negative, Neutral).
Purpose: Quickly identify conversations that need attention (negative sentiment), celebrate successful interactions (positive sentiment), or analyze patterns in user satisfaction. This is invaluable for quality assurance and customer service management.
Answer Status
Description: Filter by whether the assistant successfully answered queries (All status, Successfully Answered, No Answer Found).
Purpose: Identify knowledge gaps by filtering for "No Answer Found" messages. Use this to improve your assistant's training data and expand its capabilities. Alternatively, review successfully answered queries to understand what's working well.
Reset All Filters
Description: Click this button to clear all active filters and return to the complete, unfiltered view.
Purpose: Quickly start a new analysis without manually clearing each filter individually. This saves time when switching between different filtering criteria.
Understanding the Message List
Once you've applied your search criteria and filters, the main message list displays matching conversations. Each entry typically includes:
- Timestamp - When the conversation occurred
- User Information - Phone number or identifier of the person who messaged
- Message Preview - A snippet of the conversation content
- Status Indicators - Visual cues showing sentiment, answer status, and delivery state
- Action Buttons - Options to view full conversation details or take specific actions
Empty State
If no messages match your current filters, you'll see a "No messages found" state with the message "No messages have been sent yet." This could mean:
- No messages exist for the selected time period or filters
- Your assistant hasn't received any messages yet
- The search term didn't match any conversations
Try adjusting your filters or search terms to expand your results.
Best Practices for Using Messaging Logs
- Monitor Daily: Check the Today's Chats and Active (24h) metrics daily to stay on top of engagement levels
- Address Negative Sentiment: Regularly filter by negative sentiment to identify and resolve user issues promptly
- Identify Knowledge Gaps: Use the "No Answer Found" filter to discover topics your assistant needs to learn about
- Track Response Times: Monitor average response time to ensure your assistant is performing optimally
- Use Date Ranges: Analyze trends over different time periods to understand patterns in user behavior and engagement
- Search for Specific Topics: Use the search bar to investigate recurring questions or issues
- Combine Filters: Use multiple filters together (e.g., last 7 days + negative sentiment) for targeted analysis
- Regular Reviews: Schedule weekly or monthly reviews of messaging logs to continuously improve your assistant's performance
Troubleshooting and Analytics Tips
💡 Tip: If you notice a spike in "No Answer Found" metrics, review those conversations to identify common themes. This insight helps you prioritize updates to your assistant's knowledge base.
💡 Tip: Compare sentiment trends over different time periods to measure the impact of changes you make to your assistant's configuration or training.
💡 Tip: Use the unique users metric alongside total messages to calculate your average messages per user, helping you understand engagement depth.
Summary
The Messaging Logs feature is a powerful tool for monitoring, analyzing, and optimizing your AI assistant's performance. By leveraging the real-time analytics dashboard, smart filtering options, and sentiment analysis, you can gain deep insights into user interactions, identify areas for improvement, and ensure your assistant delivers exceptional experiences. Regular review of these logs is essential for maintaining high-quality automated communications and customer satisfaction.