Insights
Insights are saved data explorations and visualizations. When you ask the AI to analyze data, you can save successful queries as insights for reuse, sharing, and ongoing tracking.
What Are Insights
An insight captures:
- Query — The data request or analysis
- Visualization — Chart, table, or other display format
- Configuration — Display options and preferences
Once saved, an insight can be refreshed with current data anytime, maintaining the same analysis structure.
Creating Insights
Insights are created through Chat—there's no manual creation interface.
From Chat
- Go to Chat
- Ask for data analysis:
- "Show me run outcomes over the past month"
- "Create a chart of incident response times by severity"
- "What's the distribution of flow success rates?"
- The AI queries data and generates a visualization
- Review the results
- Click Save as Insight to preserve it
Refining Before Saving
Before saving, you can refine through conversation:
- "Change this to a bar chart"
- "Add a breakdown by team"
- "Filter to only show production incidents"
- "Extend the time range to 90 days"
Once satisfied, save the insight.
Browsing Insights
Navigate to Insights in the sidebar to see saved insights.
Viewing Insights
The insights list shows:
- Name and description
- Visualization preview
- Last refreshed date
- Owner
Click an insight to view full details.
Filtering
Filter by:
- My Insights — Insights you created
- Team Insights — Insights shared with your team
- Recent — Recently viewed or modified
Using Insights
Viewing an Insight
Click an insight to see:
- Full visualization
- Current data
- Configuration details
- History of refreshes
Refreshing Data
Click Refresh to rerun the query with current data. The visualization updates while maintaining the same structure and configuration.
Viewing History
Each insight tracks its refresh history:
- When it was refreshed
- Who refreshed it
- Data snapshot at that time
Managing Insights
Editing
Edit an insight to change:
- Name and description
- Visualization type
- Display options
- Sharing settings
Note: The underlying query is fixed. To change the query, create a new insight from Chat.
Sharing
Control who can see your insights:
- Personal — Only you can access
- Team — Your team can view
- Public — All users can view
Others can view and refresh shared insights but cannot edit them.
Deleting
Delete insights you no longer need. This removes the saved query and all history. Deletion cannot be undone.
Visualization Types
Insights support various visualization types:
| Type | Best For |
|---|---|
| Bar Chart | Comparing categories |
| Line Chart | Trends over time |
| Pie Chart | Proportions of a whole |
| Table | Detailed data with sorting |
| Number | Single key metric |
Changing Visualization
When creating an insight in Chat, ask for specific visualizations:
- "Show this as a line chart"
- "Create a pie chart of the distribution"
- "Display as a table with sorting"
The AI suggests appropriate visualizations based on the data, but you can override.
Alternative Suggestions
After generating a visualization, the AI may suggest alternatives:
- "This data would also work well as a line chart"
- Click suggestions to see the same data in different formats
Common Insight Examples
Operational Metrics
- "Show me flow success rates over the past 30 days"
- "Chart MTTR trends by week"
- "Display run counts by day of week"
Performance Analysis
- "Compare flow performance across teams"
- "Show top 10 most-used flows"
- "Chart guide completion rates by category"
Troubleshooting
- "Show failed runs grouped by error type"
- "Display tool execution times by tool"
- "Chart timeout frequency over time"
Best Practices
Use descriptive names that explain what the insight shows. "Q4 Flow Success Rates" is better than "Chart 1".
Include context about why this insight matters and how to interpret it.
For ongoing monitoring, refresh insights regularly to stay current.
If an insight would benefit your team, share it. Avoid creating duplicate insights for the same metrics.