Appearance
AI Insights
Get AI-powered analytics and insights about your platform performance.
Route
/admin/ai-insights
Overview
AI Insights uses artificial intelligence to analyze your platform data and provide actionable recommendations, trend analysis, and predictive insights.
Compy — Your AI Assistant
Compy is the floating AI assistant available on every admin page. Click the sparkles button in the bottom-right corner to open the chat, then ask questions about your site's analytics, competitions, orders and more in plain English.
Resizing the chat window
The button in the chat header cycles the window through three sizes — handy for longer conversations or reviewing detailed answers:
- Default — compact, tucked into the bottom-right corner.
- Large — a taller, wider window.
- Fullscreen — fills the screen (with an even margin on each side) for the most reading space.
Each click moves to the next size and then back to default. The icon and tooltip always describe the size you'll get next. Compy returns to the default size whenever you close and reopen it.
Branded avatar
Your own messages in the chat are shown with your site's favicon as the avatar, so the conversation matches your brand. If no favicon has been uploaded yet, a generic user icon is used instead — upload one on the Settings page to brand the chat.
Clearing a conversation
Once you've started chatting, the refresh button in the header clears the current conversation so you can start fresh.
Dashboard
Service Status
- AI Service Status - Online/Offline indicator
- Last Analysis - When insights were last generated
- Analysis Credits - Remaining AI analysis credits
Configuration Options
Time Period Selection
Choose the data range for analysis:
- Daily - Last 24 hours
- Weekly - Last 7 days
- Monthly - Last 30 days
- Quarterly - Last 90 days
- Yearly - Last 365 days
- Custom Range - Specify dates
Analysis Depth
Select analysis thoroughness:
Standard Analysis:
- Faster processing
- Key insights only
- Basic recommendations
- Lower credit cost
Deep Analysis:
- Comprehensive review
- Detailed recommendations
- Predictive modeling
- Trend forecasting
- Higher credit cost
Insight Categories
Performance Insights
Analyze platform performance:
- Revenue Trends - Identify growth patterns
- Conversion Rates - Optimization opportunities
- User Engagement - Activity patterns
- Competition Performance - Success metrics
User Behavior
Understand your users:
- Purchase Patterns - When and what users buy
- Engagement Times - Peak activity periods
- User Segments - Behavioral clustering
- Churn Risk - Users likely to disengage
Competition Analysis
Optimize your competitions:
- Optimal Pricing - Suggested ticket prices
- Duration Analysis - Best competition lengths
- Prize Preferences - What prizes attract entries
- Timing Recommendations - When to launch
Predictive Insights
Future forecasting:
- Revenue Projections - Expected earnings
- User Growth - Predicted user acquisition
- Competition Success - Entry predictions
- Seasonal Trends - Upcoming patterns
Report Generation
Creating a Report
- Navigate to
/admin/ai-insights - Configure time period
- Select analysis depth
- Click "Generate Insights"
- Wait for AI processing (30s - 2min)
- Review generated insights
- Export if needed
Report Contents
Each report includes:
- Executive Summary - High-level overview
- Key Findings - Most important insights
- Detailed Analysis - In-depth review
- Recommendations - Actionable suggestions
- Data Visualizations - Charts and graphs
- Trend Analysis - Pattern identification
Export Options
Download reports in multiple formats:
- PDF - Formatted report document
- CSV - Raw data for analysis
- JSON - Structured data export
- PowerPoint - Presentation slides
Recommendations
Acting on Insights
The AI provides actionable recommendations:
Pricing Optimization
- Suggested ticket price adjustments
- Bundle recommendations
- Discount strategies
Marketing Timing
- Best times to send campaigns
- Optimal launch windows
- Seasonal opportunities
User Retention
- At-risk user identification
- Re-engagement strategies
- Loyalty program suggestions
Competition Strategy
- Prize selection guidance
- Duration recommendations
- Timing optimization
Integration
Automated Reports
Schedule regular AI analysis:
- Navigate to Settings
- Enable automated insights
- Choose frequency (daily, weekly, monthly)
- Select recipients for reports
- Configure delivery method (email, dashboard)
Alert Triggers
Set up alerts for:
- Significant trend changes
- Performance anomalies
- Opportunity detection
- Risk identification
Understanding AI Confidence
Each insight includes a confidence score:
- High Confidence (80-100%) - Strong pattern, reliable recommendation
- Medium Confidence (60-79%) - Emerging pattern, consider testing
- Low Confidence (40-59%) - Weak signal, requires validation
Best Practices
Using AI Insights Effectively
- Run analysis regularly to spot trends early
- Compare different time periods
- Test recommendations on small scale first
- Combine AI insights with domain knowledge
- Track outcomes of implemented recommendations
- Use deep analysis for strategic decisions
- Use standard analysis for quick checks
Data Quality
AI insights are only as good as your data:
- Ensure accurate order tracking
- Maintain clean user data
- Track competition metrics consistently
- Monitor data completeness
Limitations
What AI Can't Do
- Guarantee specific outcomes
- Account for external factors
- Replace human judgment
- Predict black swan events
- Override business constraints
When to Use Human Judgment
- Major strategic decisions
- Ethical considerations
- Customer-facing changes
- Brand positioning
- Legal compliance
Tips
- Start with standard analysis to conserve credits
- Use deep analysis for quarterly planning
- Export reports for stakeholder meetings
- Track which recommendations worked
- Combine multiple time periods for trends
- Set up automated reports for efficiency
- Review confidence scores before acting
- A/B test AI recommendations