Analytics
SeaChat Analytics
Overview
The SeaChat Analytics feature provides valuable insights into user interactions with your AI chatbot. This dashboard enables you to track key performance metrics over time, helping measure engagement, optimize chatbot performance, and make data-driven decisions. Navigate to the Analytics tab below Agent Configuration to view the analytics page for a specific bot.
π Track Chatbot Performance Monitor the number of conversations, unique visitors, and message activity to understand how users are engaging with your chatbot.
π Measure Growth and Trends Compare data across different time ranges (Last 7 Days, Last 30 Days, etc.) to identify trends in chatbot usage and user behavior.
π€ Optimize Chatbot Responses Analyze bot message statistics to see how effectively the chatbot is handling conversations and refine responses for better user satisfaction.
π₯ Improve Customer Engagement By tracking live agent requests and user interactions, businesses can identify opportunities to enhance customer support and automation strategies.
π Data-Driven Decision Making Use real-time analytics to refine chatbot workflows, improve response accuracy, and optimize user experience based on actual usage patterns.
Settings
At the top of the page, there are a couple settings you can configure. First, select the timezone you would like the data to be presented in. Second, check the box for whether or not you would like to include conversations that have no messages from the user. For example, if a user opens the webchat and reads the welcome message, but does NOT send a message, the conversation will still be created and show up in the conversation statistics. If you check the setting box, conversations like this with no user messages will be excluded from the statistics.
By default, the dashboard data will update every hour. If you would like to refresh the data on demand, click the “Refresh Data” button.
Basic Metrics
SeaChat Analytics Basic Metrics
The Metrics section in SeaChat Analytics provides a detailed breakdown of your chatbot’s performance over a selected time range. You can customize the date range by choosing from:
- Last Day
- Last 7 Days
- Last 30 Days
- Last 90 Days
- Last 180 Days
Once a time range is selected, SeaChat will compare the chatbot’s performance during that period to the same length of time immediately before it. For example:
If you select Last 7 Days, the displayed metrics will compare the last 7 days to the previous 7-day period.
If you select Last 30 Days, it will compare the last 30 days to the 30 days before that.
Understanding the Metrics
The key performance indicators (KPIs) displayed include:
- Conversations β The total number of chatbot interactions initiated during the selected time period.
- Unique Visitors β The number of distinct users who interacted with the chatbot.
- Messages Received β The number of messages sent by users to the chatbot.
- Bot Messages Sent β The number of responses generated by the chatbot.
- Live Agent Requests β The number of times users requested a human agent.
- Agent Messages Sent β The number of messages sent by live agents in conversations.
Each metric includes a percentage change indicator, showing how it has increased or decreased compared to the previous equivalent time period. Green arrows indicate positive growth, while red arrows (if applicable) indicate a decline.
This comparison helps businesses track chatbot performance trends over time, evaluate improvements, and identify areas for optimization.
Conversation Yearly Overview
Conversation Yearly Overview
The Conversation Overview section in SeaChat Analytics provides a high-level summary of chatbot interactions for a selected year. Users can choose a specific year from the dropdown menu to view key statistics on chatbot engagement during that period.
- Total Number of Conversations β Displays the total count of chatbot conversations for the selected year. Average Number of Messages Per Conversation β Shows the average number of messages exchanged per conversation, helping assess user engagement levels.
- Total Messages by Month
Below the key metrics, a line graph visualizes the total number of messages exchanged with the chatbot each month. This allows users to:
- Identify trends in chatbot activity over time.
- Spot seasonal fluctuations or spikes in engagement.
- Assess the impact of business campaigns or chatbot improvements on conversation volume.
Conversation & Messages Breakdown
The Conversations Breakdown by Day, Time & Channel section allows users to analyze chatbot interactions across various channels and time periods within a custom date range. By selecting a specific start and end date, businesses can gain detailed insights into when and where most of their interactions are occurring.
Channel Breakdown
The channel breakdown table displays the following data for each channel:
- Channel β The different communication channels through which users are interacting with the chatbot (e.g., Webchat, LINE, WhatsApp, Voice, Messenger, etc.). All channels offered by SeaChat will automatically appear in the table once they have traffic.
- Unique Visitors β The number of distinct users who engaged with the chatbot via each specific channel within the selected time frame.
- Inbound Messages Received β The total number of messages sent by users across each channel during the selected date range.
Setting Up Sub-Channel Tracking
The SeaChat WebChat widget allows you to install a chat window directly in your own website.
Some clients add the webchat widget to multiple sites and want to be able to track the amount of traffic from the widget on each site separately.
By default, the channel breakdown will lump all the webchat traffic together into a single channel type called WEBCHAT.
However, with a simple customization to the WebChat widget code, you can provide a specific “subchannel” name to each instance of the widget and track the traffic separately.
Once you customize the webchat widget with subchannel information, all subsequent traffic will appear as WEBCHAT - {subchannel} in the table.
WebChat Subchannel Setup
Navigate to Channels -> WebChat -> Install Widget to find the WebChat widget code.
Within this code block, there are four instances of your chatbot’s webchat URL - it looks like this: https://chat.seasalt.ai/chat/{chat_config_id}.
To distinguish different subchannels, simply append to following to the end of the URL: ?channel={subchannel_name}.
For example, say you want to add the WebChat widget to your homepage as well as your wiki site.
You could update the URL https://chat.seasalt.ai/chat/aaaabbbbccccdddd to https://chat.seasalt.ai/chat/aaaabbbbccccdddd?channel=homepage and add the widget code to your homepage.
Then you could update the URL again to https://chat.seasalt.ai/chat/aaaabbbbccccdddd?channel=wiki and add the code to your wiki site.
In the channel breakdown, you would see traffic from two separate channels: WEBCHAT - homepage and WEBCHAT - wiki.
Day of the Week Breakdown
The ‘Day of the Week’ breakdown allows you to see how many user messages are sent to your bot on each day of the week. This allows you to identify trends of when users are engaging with your bot.
Hour of the Day Breakdown
The ‘Hour of the Day’ breakdown allows you to see what time of day users are sending messages to your bot. This allows you to identify trends of when users are engaging with your bot, and may help with making decisions such as what times to have live agents online.