Analytics & Reporting¶

Archivus provides comprehensive analytics and compliance reporting powered by DuckDB for 10-100x faster aggregations. Export to PowerBI, Tableau, or custom dashboards.

Analytics Architecture¶

Team+

Hybrid Analytics:

PostgreSQL: Operational data (OLTP)
DuckDB/MotherDuck: Analytics data (OLAP)
Parquet on S3: Long-term storage

Benefits:

10-100x faster analytics queries
No impact on operational performance
Cost-effective long-term retention
Standard SQL interface

Dashboard Overview¶

Real-Time Metrics:

Document velocity (uploads per period)
Active users and top contributors
Storage consumption trends
AI credit usage by operation
Agent performance metrics
Search analytics and patterns

Document Analytics¶

Team+

Upload Trends¶

Track document growth over time:

Documents per day/week/month/quarter
Upload patterns and seasonality
Growth rate and projections
Storage capacity forecasts

Visualizations:

Time series charts
Growth trend lines
Seasonal patterns
Capacity planning

Document Distribution¶

Understand your document portfolio:

Documents by type (invoice, contract, etc.)
Documents by workspace/project
Documents by owner/contributor
File format distribution

Use Cases:

Identify document classification patterns
Optimize storage allocation
Plan automation priorities

Processing Metrics¶

Monitor document processing pipeline:

Average processing time
Success/failure rates
Retry and error rates
Bottleneck identification

User Analytics¶

Team+

Activity Tracking¶

Monitor user engagement:

Active users (daily/weekly/monthly)
Top contributors by uploads
Most active by searches
Dormant account identification

Collaboration Metrics¶

Measure team collaboration:

Documents shared between users
Comment activity
@mention frequency
Workspace participation

Adoption Analysis¶

Track feature adoption:

AI tool usage by user
Search type preferences
Workflow utilization
MCP integration activity

AI Usage Analytics¶

Team+

Credit Tracking¶

Real-time credit monitoring:

Current balance
Credits used this period
Credits by operation type
Projected runout date

Alerts:

75% usage warning
90% usage alert
Low balance notifications
Anomaly detection

Operation Breakdown¶

Understand AI usage patterns:

Credits by operation type (classify, summarize, extract, etc.)
Cost per document processed
Model selection distribution (Claude, Gemini, OpenAI)
Token usage and cost analysis

Optimization Insights:

Expensive operations identification
Cost reduction opportunities
Model selection recommendations
Batch processing suggestions

ROI Analysis¶

Calculate AI automation value:

Time saved per operation type
Labor cost avoided
Efficiency gains
Productivity improvements

Example ROI Report:

AI Automation ROI - January 2026
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Documents Processed:      1,247
AI Credits Used:          3,845
Average Time Saved:       12 min/doc
Total Time Saved:         249 hours

Labor Rate:               $50/hour
Labor Cost Avoided:       $12,450
AI Credit Cost:           $385
Net Savings:              $12,065

ROI:                      3,131%

Agent Analytics¶

Pro+

Track AI agent performance:

Performance Metrics:

Total evaluations performed
Auto-approval rate
Average confidence scores
Time saved estimates
Actions proposed vs executed

Quality Metrics:

User override rate
Correction frequency
Pattern accuracy
Learning progression

Dashboard View:

Classification Agent - 30 Days
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Evaluations:          847
Auto-Approved:        812 (95.9%)
User Overrides:       35 (4.1%)
Avg Confidence:       93.7%
Time Saved:           ~28 hours

Top Classifications:
  1. INVOICE         347 (41%)
  2. CONTRACT        198 (23%)
  3. RECEIPT         142 (17%)
  4. REPORT          87 (10%)
  5. OTHER           73 (9%)

Search Analytics¶

Team+

Query Patterns¶

Understand search behavior:

Most common queries
Search frequency by user
Semantic vs text search ratio
Result click-through rates

Performance Metrics¶

Monitor search effectiveness:

Average response time
Result relevance scores
Searches with no results
Refine/retry patterns

Content Gaps¶

Identify documentation needs:

Common searches with poor results
Missing document types
Under-tagged content
Knowledge base gaps

Compliance Reporting¶

Team+

Audit Trails¶

Export comprehensive audit logs:

Document access events
User actions (view, download, edit, delete)
AI operations performed
Admin actions
Agent automation history

Export Formats:

CSV for analysis
XLSX with formatting
PDF for presentation
JSON for programmatic access

Retention:

Standard: 90 days
Team: 1 year
Enterprise: 2 years (configurable)

Compliance Reports¶

Pre-built compliance templates:

HIPAA: PHI access tracking
GDPR: Data processing logs
SOC 2: Security event audit
ISO 27001: Information security
SEC: Document retention compliance

Custom Dashboards¶

Enterprise

PowerBI Integration¶

Live data feed to PowerBI:

Direct SQL connection to DuckDB
Real-time or scheduled refresh
Custom visualizations
Executive dashboards

Connection String:

Server: duckdb.archivus.app
Database: tenant_{uuid}
Authentication: API Key

Tableau Connector¶

Native Tableau integration:

Web data connector
Automatic schema detection
Incremental refresh support
Published data sources

Custom API¶

Build custom analytics:

# Get document trends
GET /api/v1/analytics/documents/trends
  ?start_date=2026-01-01
  &end_date=2026-12-31
  &group_by=month
  &organization_id=org-uuid

# Get AI usage breakdown
GET /api/v1/analytics/ai/usage
  ?period=month
  &breakdown=operation
  &format=json

# Get agent performance
GET /api/v1/analytics/agents/performance
  ?agent_id=agent-uuid
  &period=quarter

Export Options¶

CSV Export:

Raw data for analysis
All metrics available
Timestamp precision
UTF-8 encoding

Excel Export:

Formatted tables
Charts and graphs
Multiple worksheets
Conditional formatting

PDF Reports:

Professional formatting
Charts and visualizations
Custom branding
Scheduled delivery

Scheduled Reports¶

Team+

Automate report generation:

Schedule Options:

Daily, weekly, monthly, quarterly
Specific day/time
After workflow completion
On-demand via API

Delivery Methods:

Email attachment
Slack notification
Webhook POST
S3/storage upload

Example Scheduled Report:

report:
  name: Weekly Executive Summary
  schedule: "0 9 * * MON"  # Every Monday 9am
  recipients:
    - executives@company.com
  format: pdf
  sections:
    - document_velocity
    - ai_credit_usage
    - top_contributors
    - compliance_summary
  branding:
    logo: company-logo.png
    colors: ["#1E40AF", "#3B82F6"]

Department Chargeback¶

Enterprise

Allocate costs to departments:

Cost Attribution:

AI credits by organization/workspace
Storage by department
API usage by application
User activity by cost center

Chargeback Report:

Department Cost Allocation - January 2026
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Legal Department:
  AI Credits:       12,450 ($1,245)
  Storage:          125 GB ($25)
  Total:            $1,270

Finance Department:
  AI Credits:       8,920 ($892)
  Storage:          78 GB ($15.60)
  Total:            $907.60

Operations:
  AI Credits:       5,340 ($534)
  Storage:          45 GB ($9)
  Total:            $543

Getting Started¶

Navigate to Analytics dashboard
Select date range and metrics
Apply filters (workspace, user, type)
Export or schedule reports
Integrate with BI tools (Enterprise)

View Analytics API Docs → See Report Examples →