Documentation
Learn how to read and interpret your analytics dashboard
Overview
Welcome to your analytics platform! This dashboard helps you understand how your customers interact with your business through various communication channels.
What You Can Do
- Track customer Messages across multiple channels
- Understand customer sentiment (positive, negative, neutral)
- Identify peak usage times for better resource allocation
- Analyze conversation trends over time
- Export reports for sharing with your team
Dashboard Layout
Your dashboard is organized into several key areas:
Header Section
Contains date range selector and template/account switcher. Use this to filter data for specific time periods or business units.
Metric Cards
Quick summary cards showing key performance indicators (KPIs) like total Messages, conversations, and engagement rates at a glance.
Charts & Visualizations
Interactive charts that help you visualize trends, patterns, and insights from your data.
Messages
What it means: A Message is a one complete exchange between your business and a customer through any channel (chat, email, social media, etc.).
Example
If a customer messages you asking about store hours, and you respond with the information, that's one Message.
Why It Matters
- Higher Message counts mean more customer engagement
- Track if your marketing campaigns are driving more interactions
- Identify which channels customers prefer to use
Conversations
What it means: A conversation represents a period of continuous activity. It includes all Messages and interactions within a specific timeframe before the user becomes inactive.
Example
A customer visits your website chat, asks 3 different questions over 5 minutes. That's 1 conversation with 3 Messages.
Conversation vs Message
Think of a conversation as a visit to you, and a Message as the different questions asked during that visit.
Engagement Rate
What it means: The average number of Messages in Conversations (replies, actions taken, etc.).
How It's Calculated
A 8 engagement rate means on an average people send 8 messages per conversation.
What's a Good Rate?
- • Above 5 - Excellent engagement
- • 3-5 - Good engagement
- • Below 3 - Room for improvement
NOTE: This is a general guideline and individual results may vary. for example on our SmartKiosk 2 messages per conversation might be typical.
Line Charts
What it shows: Line charts display trends over time, making it easy to see if things are going up, down, or staying steady.
When to Use
- Tracking daily, weekly, or monthly conversation volume
- Comparing current period vs previous period
- Identifying seasonal patterns or trends
How to Read It
- • X-axis (horizontal): Time (days, weeks, months)
- • Y-axis (vertical): Number of Messages/conversations
- • Line going up: Increasing activity
- • Line going down: Decreasing activity
- • Flat line: Stable activity
Bar Charts
What it shows: Bar charts compare different categories side-by-side, making it easy to see which one is highest or lowest.
When to Use
- Comparing Messages across different channels (Instagram, Facebook, etc.)
- Showing top conversation topics or categories
- Comparing performance across different templates or accounts
How to Read It
The taller the bar, the higher the value. Quickly spot your best and worst performing categories by comparing bar heights.
Pie Charts
What it shows: Pie charts show proportions or percentages of a whole, like slices of a pie.
When to Use
- Showing sentiment distribution (positive, negative, neutral)
- Breaking down Messages by category or topic
- Displaying channel usage percentages
How to Read It
Each slice represents a portion of the total. Bigger slices = larger percentage. All slices together = 100%.
Heatmaps
What it shows: Heatmaps use colors to show intensity or concentration, like a weather map showing temperatures.
When to Use
- Identifying peak usage hours (which days/times are busiest)
- Spotting patterns in customer behavior
- Planning staff schedules based on busy periods
How to Read It
- • Dark/Hot colors (red, orange): High activity
- • Light/Cool colors (blue, green): Low activity
- • Look for patterns: Are weekends busier? Morning or evening rush?
Sentiment Analysis
What it means: Sentiment analysis automatically determines if customer messages are positive, negative, or neutral based on the words they use.
😊 Positive Sentiment
Customers expressing satisfaction, happiness, or gratitude. Example: "Thank you so much! This was exactly what I needed."
😐 Neutral Sentiment
Factual questions or statements without strong emotion. Example: "What are your store hours?"
😞 Negative Sentiment
Customers expressing frustration, disappointment, or complaints. Example: "I'm very disappointed with this service."
How to Use This Data
- High negative sentiment? Investigate what's causing customer frustration
- Track sentiment trends over time to measure service improvements
- Prioritize responding to negative sentiment Messages first
Intent Classification
What it means: Intent classification identifies why a customer is reaching out - what they're trying to accomplish in their message.
🔍 Inquiry
Customer is asking a question or seeking information.
Example: "What time do you close?" or "Do you ship internationally?"
⚠️ Complaint
Customer is expressing dissatisfaction or reporting a problem.
Example: "My order arrived damaged" or "The service was very slow"
💬 Feedback
Customer is sharing their opinion, suggestion, or review.
Example: "I loved the new menu items!" or "You should add more sizes"
📝 Request
Customer is asking for a specific action or service.
Example: "Can I get a refund?" or "Please send me a receipt"
Why Intent Matters
- Complaints need urgent attention to prevent customer churn
- Inquiries can often be automated with FAQ or chatbot responses
- Feedback provides valuable insights for product/service improvements
- Requests may require specific staff permissions or actions
Message Categories
What it means: Categories group messages by topic or subject matter, helping you understand what your customers are talking about.
👕 Apparel
Messages about clothing, fashion items, sizes, fits, or style recommendations.
🍔 Food
Questions or discussions about menu items, ingredients, dietary restrictions, or dining options.
💻 Technology
Technical questions, device issues, software problems, or IT-related inquiries.
🎭 Leisure
Entertainment, recreational activities, hobbies, or relaxation-related topics.
⚽ Sports
Athletic activities, sports equipment, fitness programs, or game-related discussions.
🎉 Events
Special occasions, bookings, reservations, conferences, or scheduled activities.
🏨 Amenity
Facility features, services provided, accommodation details, or property-related questions.
🌐 Language
Messages about translations, language support, or communication in different languages.
❓ Undefined
Messages that don't fit into specific categories or need manual classification.
🚫 Abusive
Messages containing inappropriate language, harassment, or violations of community guidelines.Requires immediate attention and moderation.
Using Categories Effectively
- Route messages to specialized team members based on category
- Identify trending topics to inform business decisions
- Create category-specific training materials for staff
- Track which categories have highest volume or longest response times
- Flag Abusive messages for immediate review and action
Conversation Trends
What it shows: How your conversation volume changes over time and compares to previous periods.
Understanding Trend Indicators
- • ↗ Upward trend: Messages are increasing (good for growth!)
- • ↘ Downward trend: Messages are decreasing (investigate why)
- • → Flat trend: Stable conversation volume
What to Look For
- Sudden spikes: Did you run a marketing campaign or promotion?
- Sudden drops: Technical issues? Change in marketing strategy?
- Seasonal patterns: Back-to-school rush? Holiday shopping?
Category Breakdown
What it shows: Groups Messages into topics or categories based on what customers are asking about.
Common Categories
How This Helps
- Identify which topics customers care about most
- Create FAQ content for frequently asked questions
- Allocate staff training based on common inquiry types
- Improve self-service resources for top categories
Peak Usage Times
What it shows: When your customers are most active, broken down by day of week and time of day.
Example Insights
Your heatmap might show that Mondays at 9 AM and Fridays at 5 PM are your busiest times, while weekend mornings are quiet.
How to Use This
- Schedule more staff during peak hours
- Set up automated responses during off-peak times
- Plan system maintenance during low-traffic periods
- Launch campaigns when your audience is most engaged