​

Metrics & Analytics

Backstack provides comprehensive metrics and analytics to help you understand how your organization uses AI tools, track performance, and optimize your AI infrastructure.

​

Overview

The Analytics dashboard gives you visibility into:

• Tool Execution - Which tools are being used and how often
• Performance - Execution times and success rates
• User Activity - Who is using which tools and workspaces
• Client Distribution - Which AI clients are connecting
• Trends Over Time - Usage patterns and growth

​

Key Metrics

​

Tool Execution Metrics

Total Executions:

• Number of times tools have been called
• Breakdown by tool name
• Success vs failure counts
• Execution frequency over time

Most Used Tools:

• Ranking of tools by execution count
• Percentage of total tool usage
• Trends (increasing or decreasing usage)
• Comparison across workspaces

Execution Success Rate:

• Percentage of successful tool calls
• Failure reasons and error patterns
• Tools with highest failure rates
• Opportunities for configuration improvements

​

Performance Metrics

Execution Time:

• Average execution time per tool
• Min, max, and median execution times
• Slow tool identification
• Performance trends over time

Response Latency:

• Time from request to first response
• AI provider response times
• End-to-end conversation latency
• Performance by workspace and time of day

Resource Usage:

• Number of active AI conversations
• Documents indexed and searched
• Desktop devices connected
• Storage consumption

​

User Activity Metrics

Active Users:

• Daily active users (DAU)
• Weekly active users (WAU)
• Monthly active users (MAU)
• User growth trends

User Engagement:

• Tools used per user
• Conversations per user
• Documents shared per user
• Average session duration

Top Users:

• Users with most tool executions
• Power users and early adopters
• Users needing additional training
• User adoption by department or team

​

Workspace Metrics

Workspace Activity:

• Tool executions by workspace
• Active members per workspace
• Documents shared to each workspace
• Most active workspaces

Workspace Adoption:

• Percentage of users per workspace
• Workspace creation trends
• Workspace size distribution
• Underutilized workspaces

​

Client Distribution

AI Client Usage:

• Claude Desktop vs VS Code vs Cursor, etc.
• Client versions in use
• Platform distribution (Windows, macOS, Linux)
• Client adoption trends

Connection Health:

• Active connections
• Connection failures and errors
• Average uptime per client type
• Authentication success rates

​

Viewing Analytics

​

Accessing the Dashboard

1. Navigate to Organization → Analytics
2. View the default dashboard with key metrics
3. Use date range selector to adjust timeframe
4. Apply filters to drill down into specific metrics

​

Time Range Selection

Choose the period to analyze: Preset Ranges:

• Last 24 Hours
• Last 7 Days
• Last 30 Days
• Last 90 Days
• All Time

Custom Range:

• Select specific start and end dates
• Compare different time periods
• Analyze seasonal patterns
• Measure before/after changes

​

Filtering Options

By Workspace:

• Select one or multiple workspaces
• Compare workspace performance
• Identify workspace-specific patterns
• Optimize per-workspace tool allocation

By Tool:

• Focus on specific tools
• Track tool adoption over time
• Identify underutilized tools
• Plan tool retirement or promotion

By User:

• Individual user analytics
• User cohort comparison
• Training effectiveness measurement
• Power user identification

By Client:

• Claude Desktop activity
• VS Code extension usage
• Cursor integration metrics
• Compare client effectiveness

​

Analytics Views

​

Executive Dashboard

High-level overview for leadership: Key Performance Indicators (KPIs):

• Total tool executions this month
• Active users this week
• Average success rate
• Month-over-month growth

Trend Graphs:

• Daily tool execution volume
• User adoption curve
• Tool success rate over time
• Resource consumption trends

Top Lists:

• Top 10 most-used tools
• Top 10 active users
• Top 5 workspaces by activity
• Most common failure reasons

​

Tool Analytics

Deep dive into tool usage: Tool Performance Table: Execution Timeline:

• Line graph showing executions over time
• Identify peak usage hours
• Spot usage anomalies
• Plan capacity based on patterns

Error Analysis:

• Common error messages
• Tools with highest failure rates
• Root cause patterns
• Suggested fixes

​

User Analytics

Understand how your team uses AI: User Activity Heatmap:

• Usage by day of week and hour
• Identify peak times
• Optimize support coverage
• Plan maintenance windows

Adoption Funnel:

1. Users invited to organization
2. Users who completed first login
3. Users who executed first tool
4. Users active in last 7 days
5. Power users (daily usage)

User Segmentation:

• Active, inactive, and dormant users
• Users by department or role
• Early adopters vs late majority
• Training needs by segment

​

Workspace Analytics

Optimize workspace organization: Workspace Comparison:

• Side-by-side workspace metrics
• Identify best-performing workspaces
• Replicate success patterns
• Consolidate underutilized workspaces

Member Activity:

• Active vs inactive members per workspace
• Member contribution distribution
• Collaboration patterns
• Optimal workspace size

Tool Assignment:

• Which tools are assigned to which workspaces
• Tool utilization per workspace
• Unused tool assignments
• Opportunities to share tools across workspaces

​

Use Cases

​

Capacity Planning

Monitor growth trends:

• Track daily/weekly execution volume
• Predict future resource needs
• Plan infrastructure scaling
• Budget for AI provider costs

Example:

• Current: 10,000 tool executions/day
• Growth rate: 15% month-over-month
• Projected need in 6 months: 20,000/day
• Action: Plan capacity increase or optimize usage

​

Tool Optimization

Identify underperforming tools:

• Tools with low success rates
• Tools taking too long to execute
• Tools with configuration issues
• Tools not being used despite availability

Example:

• PostgreSQL server shows 60% failure rate
• Investigation reveals missing environment variables
• Fix: Update DATABASE_URL for all instances
• Result: Success rate improves to 98%

​

User Training

Measure adoption and effectiveness:

• Which users need training
• Which tools users struggle with
• Training program effectiveness
• Knowledge gaps by team

Example:

• New users execute 80% fewer tools than experienced users
• Action: Create onboarding program
• Measure: Track new user adoption rate
• Result: New users reach productivity 50% faster

​

Cost Management

Track AI provider costs:

• Total tool executions (proxy for AI costs)
• Breakdown by workspace and user
• Identify cost optimization opportunities
• Set usage budgets and alerts

Example:

• Development workspace uses 60% of tool executions
• Analysis shows inefficient workflow
• Action: Optimize development scripts
• Result: 30% reduction in executions

​

Security Monitoring

Detect unusual patterns:

• Spike in tool executions from single user
• Unusual tool access patterns
• Off-hours activity
• Execution patterns violating policies

Example:

• User executes 1000 file deletion tools in 5 minutes
• Alert triggers for investigation
• Discovery: User was testing script, not malicious
• Action: Add rate limiting to prevent accidents

​

Exporting Analytics

​

Download Reports

Export metrics for offline analysis:

1. Select metrics and filters
2. Choose date range
3. Click Export
4. Select format:
   • CSV - For spreadsheet analysis
   • PDF - For presentations and reports
   • JSON - For programmatic processing

Export Contents:

• Raw metric data
• Aggregated statistics
• Time-series data
• Filter parameters applied

​

Scheduled Reports

Automate report generation:

1. Navigate to Analytics → Scheduled Reports
2. Click Create Schedule
3. Configure:
   • Report type (Executive, Tool, User, Workspace)
   • Metrics to include
   • Frequency (daily, weekly, monthly)
   • Recipients
   • Format
4. Save schedule

Reports are generated and delivered automatically.

​

API Access

For advanced analytics:

• Export data via GraphQL API
• Build custom dashboards
• Integrate with business intelligence tools
• Create real-time monitoring

See API documentation for endpoints and authentication.

​

Best Practices

​

Regular Review

Daily:

• Check tool execution volume
• Monitor error rates
• Review security alerts

Weekly:

• User activity trends
• Tool adoption progress
• Top user and workspace analysis

Monthly:

• Executive dashboard review
• Capacity planning check
• Cost analysis
• Training needs assessment

Quarterly:

• Long-term trend analysis
• Tool portfolio review
• User satisfaction measurement
• Strategic planning based on metrics

​

Actionable Insights

Don’t just collect data - act on it:

• High failure rate? → Investigate configuration
• Low adoption? → Improve training or tool selection
• Spike in usage? → Plan capacity increase
• Unused tool? → Remove or promote to right users
• Power user identified? → Make them a champion

​

Benchmarking

Compare across dimensions:

• Workspace A vs Workspace B
• This month vs last month
• Your org vs industry averages (if available)
• Tool X vs Tool Y effectiveness

​

Goal Setting

Use metrics to track progress toward goals: Example Goals:

• Increase active users by 20% this quarter
• Achieve 95% tool success rate
• Reduce average execution time by 30%
• Reach 100% user adoption by year-end

Track metrics against goals monthly and adjust strategies.

​

Troubleshooting

​

Metrics Not Updating

Problem: Dashboard shows stale data Solutions:

• Refresh the page
• Check date range includes current time
• Verify you have permission to view metrics
• Wait a few minutes for metrics aggregation
• Contact support if data is >1 hour old

​

Missing Data

Problem: Expected metrics don’t appear Solutions:

• Verify events occurred during selected date range
• Check filters aren’t excluding data
• Ensure tools were actually executed (check activity logs)
• Confirm workspace/user has activity

​

Performance Issues

Problem: Analytics dashboard loads slowly Solutions:

• Reduce date range
• Apply more specific filters
• Export data for offline analysis
• Use preset views instead of custom
• Contact support for large datasets

​

Export Fails

Problem: Report export doesn’t complete Solutions:

• Reduce date range or metric scope
• Try different format
• Check browser allows downloads
• Use scheduled reports for large exports
• Contact support for assistance

​

Privacy Considerations

What’s tracked:

• Tool execution counts and timing
• User activity patterns
• Workspace usage statistics
• Success/failure rates

What’s NOT tracked:

• Conversation content or message text
• Personal data in tool parameters
• Document contents
• API keys or credentials

Metrics are aggregated and anonymized where appropriate to protect user privacy while providing useful insights.

​

Next Steps