The Insights Overview offers detailed data on your chat's performance over a specific time period. It highlights issues with your workflow and suggests ways to improve it. Thus, by using the Insights Overview, you can increase customer satisfaction, address and improve gaps in information, and gain a deeper understanding of your customers' behavior.
Access Insights Overview
- Click Solve > Widget Insights > Overview.
Chat performance
The Chat performance provides comprehensive data, covering various key metrics about the chat’s overall effectiveness:
-
Total Chats
This counts the number of times a user opens a chat widget. It will be deemed a single chat session if the user revisits the widget within 3 hours of the last interaction. If the user restarts the conversation or refreshes the browser tab within 3 hours, it will only count as 1 chat session. -
Deflections
Counts chat sessions where the user was not transferred to a human agent (handoff) or did not issue a ticket. Even if a single chat session involved multiple handoffs or generated several tickets, it counts as one non-deflected chat. -
Average CSAT
The average Customer Satisfaction (CSAT) scores from chats. -
Relevance (only visible when Discover is enabled)
Uses AI to rate how well the response addresses the user's inquiry. Each deflected chat receives one relevance rating. Non-deflected chats and some historical chats won’t have a relevance rating. -
Engagement (only visible when Discover is enabled)
Uses AI to determine if the user stayed engaged throughout the chat or dropped off midway. -
Quick Feedback
Direct feedback from users about the usefulness of the article when it was surfaced. -
Realized Savings
This is dollar savings based on the configuration in your organization.
Insights card
The Insights Card provides essential information about your chat performance. It highlights areas that need attention and offers suggestions for improvement. Your dashboard displays three different insight cards, but some may not be visible if Discover is not enabled in your Forethought dashboard.
1. Worst performing workflow
This displays the top three workflows with the highest number of non-deflected chats, pinpointing workflows that need immediate attention. By reviewing where customers frequently drop off and enhancing these workflows, you can improve both self-service rates and customer satisfaction. It counts all conversations across all workflows and shows the top three workflows with the highest non-deflection rates for the given date range.
2. Most cost-inefficient chat topics
This card is only visible to customers with Discover. It shows the top three chat topics with the most potential savings. To maximize these potential savings, we recommend creating workflows or knowledge articles based on the performance of existing workflows, agent responses, and identified knowledge gaps.
3. Most unhappy topics
This is only visible to customers with Discover. This section highlights the top three chat topics with the lowest Customer Satisfaction Survey scores.
Chat trends
This includes comprehensive data breakdowns over specific periods: weekly (last 4 weeks), monthly (last 4 months), and quarterly (last 4 quarters). The key metrics included are the following:
- Chats
- Deflection rate
- Knowledge articles surfaced
- Average CSAT score
- Quick feedback
- Relevance (only visible when Discover is enabled)
- User engagement (only visible when Discover is enabled)
Chat comparison
This section compares various performance metrics with "Chat Performance" for a selected time frame and includes data from the previous period for easy comparison. The chat comparison section presents the following graphs:
-
- Time Series Comparison: Displays historical data of multiple metrics in an easy-to-read line graph. Each metric includes data from the previous period, providing context for identifying trends.
- Deflections: Shows the number of deflections and non-deflections during a period, along with potential and realized savings.
- CSAT: Displays the average Customer Satisfaction (CSAT) score of all chats, including the total number of responses and the breakdown of responses by rating.
- Relevance: Shows the number of relevant, somewhat relevant, and irrelevant chats. The relevance rating uses AI to determine how well the response directly addresses the user's query. Non-deflected chats and some historical chats won't have a relevance rating.
- Quick Feedback: Displays the percentage of positive feedback from users, the total number of responses, and the number of users who did not provide feedback when articles surfaced.
- Engagement: Shows the percentage of chats where users were engaged. It also displays the number of chats where users were engaged throughout the chat or dropped off midway.
When you hover over a specific data point, a mini popup will display:
-
- Date range
- Volume (as a number) or value (as a percentage)
- Breakdown of all enabled metrics