docs: add comprehensive SaaS product requirements documentation

Add complete documentation for Wareflow SaaS 1.0.0 including system architecture,
data models, import plugin system, and MVP roadmap. Documents cover the vision for
a native warehouse analysis platform with WMS integration via transform plugins.

This establishes the foundation for rebuilding the system from scratch as a true
SaaS with:
- Import plugin system for WMS-specific data transformation
- Rich normalized data model with full traceability (who/when/where/why)
- Extensible TypeScript analysis engine
- Template-driven workflows
- Professional Excel export

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Author: codewizdave

docs/saas/01-overview.md

@@ -0,0 +1,223 @@
+# Product Overview
+
+## Executive Summary
+
+Wareflow SaaS is a **warehouse analysis platform** that transforms raw Excel data into actionable insights through automated analysis and reporting. Unlike traditional warehouse management systems, Wareflow focuses specifically on **post-facto analysis** of warehouse operations.
+
+## Problem Statement
+
+Warehouse managers face three fundamental challenges:
+
+1. **Data Fragmentation**: Critical data locked in disparate Excel files with inconsistent formats
+2. **Analysis Complexity**: No easy way to analyze operations without technical expertise or expensive WMS
+3. **Reporting Burden**: Time-consuming manual report generation with inconsistent formatting
+
+## Solution
+
+Wareflow provides a **template-driven, extensible analysis platform** that:
+- Ingests raw Excel files through intelligent mapping
+- Automatically unlocks analyses based on available data
+- Executes analyses using a modular TypeScript engine
+- Exports professional Excel reports with one click
+- Extends functionality through a plugin system
+
+## Target Users
+
+### Primary Users
+
+| Role | Needs | Pain Points |
+|------|-------|-------------|
+| **Warehouse Managers** | Operational insights, team performance | Data in spreadsheets, manual reporting |
+| **Operations Analysts** | Deep-dive analysis, trend identification | Complex Excel formulas, repetitive work |
+| **Supply Chain Directors** | Strategic overview, KPI tracking | No visibility across warehouses |
+
+### Secondary Users
+
+| Role | Needs | Pain Points |
+|------|-------|-------------|
+| **IT/Data Teams** | Data integration, automation | Custom scripts, maintenance burden |
+| **Finance Teams** | Cost analysis, ROI calculations | Manual data aggregation |
+
+## Core Value Propositions
+
+### 1. Time to Insight
+- **5 minutes** from raw Excel to first analysis
+- Automatic analysis unlocking based on data
+- No SQL or programming required
+
+### 2. Extensibility
+- Add new analyses without recompilation
+- Plugin system for custom analyses
+- Open-source analysis library
+
+### 3. Data Ownership
+- Local-first data storage option
+- Full data export capability
+- No vendor lock-in
+
+### 4. Professional Output
+- Publication-ready Excel reports
+- Consistent formatting
+- Customizable templates
+
+## Product Philosophy
+
+### Data-First Approach
+
+```
+Raw Data → Immediate Access → Analysis → Reporting
+   ↑                                      ↓
+   └─────────── Foundation ────────────────┘
+```
+
+The system prioritizes **data accessibility** above all else:
+- Raw data must be queryable immediately after import
+- No "black box" transformations
+- Full transparency in data processing
+
+### Template-Driven Workflows
+
+Templates bridge the gap between raw data and analysis:
+
+```typescript
+Template "Basic Warehouse"
+├── Required Tables: produits, mouvements
+├── Enabled Analyses:
+│   ├── ABC Classification
+│   └── Inventory Overview
+└── Validation Rules:
+    ├── Primary key uniqueness
+    └── Date format validation
+```
+
+### Extensibility by Design
+
+Every component is designed for extension:
+- **Analyses**: Add via plugin registration
+- **Templates**: Create custom schemas
+- **Export Formats**: Define new output templates
+- **Data Sources**: Implement new import adapters
+
+## Key Differentiators
+
+### vs. Traditional WMS
+
+| Feature | Wareflow SaaS | Traditional WMS |
+|---------|---------------|-----------------|
+| Focus | Analysis only | Operations mgmt |
+| Implementation | 5 minutes | Months |
+| Cost | Low | Very high |
+| Complexity | Low | Very high |
+
+### vs. Business Intelligence Tools
+
+| Feature | Wareflow SaaS | BI Tools |
+|---------|---------------|----------|
+| Learning Curve | Minimal | Steep |
+| Warehouse Domain | Pre-built | Generic |
+| Excel Integration | Native | Complex |
+| Time to First Report | Minutes | Days/Weeks |
+
+### vs. Spreadsheets
+
+| Feature | Wareflow SaaS | Excel Only |
+|---------|---------------|------------|
+| Data Validation | Automatic | Manual |
+| Analyses | Pre-built | Manual formulas |
+| Consistency | Guaranteed | Error-prone |
+| Scalability | High | Limited |
+
+## Success Metrics
+
+### Product Metrics
+- **Time to First Analysis**: < 5 minutes
+- **Template Matching Rate**: > 90% auto-match
+- **Analysis Execution Time**: < 10 seconds (10K rows)
+- **Data Import Success Rate**: > 95%
+
+### Business Metrics
+- **User Retention**: > 80% after 30 days
+- **Weekly Active Users**: > 60%
+- **Report Generation**: > 10 reports/user/week
+- **Analysis Extension**: > 20% of users add custom analyses
+
+## Non-Goals
+
+### Out of Scope for v1.0
+
+1. **Real-time Operations**
+   - No live warehouse operations management
+   - No barcode scanning integration
+   - No IoT/sensor integration
+
+2. **Multi-warehouse Orchestration**
+   - No cross-warehouse transfers
+   - No centralized inventory management
+
+3. **Predictive Analytics**
+   - No ML-based demand forecasting
+   - No prescriptive recommendations
+   - Focus on descriptive and diagnostic analytics
+
+4. **Collaboration Features**
+   - No multi-user editing
+   - No comment/annotation system
+   - No dashboard sharing
+
+### Future Considerations
+
+These may be added in future versions:
+- Web-based interface (currently desktop-first)
+- Mobile companion app
+- API for integrations
+- Advanced anomaly detection with ML
+
+## Design Principles
+
+### 1. Simplicity Over Complexity
+- Common use cases should be simple
+- Power features available but not intrusive
+- Progressive disclosure of complexity
+
+### 2. Transparent Data Processing
+- Show every transformation step
+- Explain analysis results
+- Provide data lineage
+
+### 3. Fail Gracefully
+- Validate before processing
+- Meaningful error messages
+- Recovery suggestions
+
+### 4. Performance at Scale
+- Virtual scrolling for large datasets
+- Lazy loading of analyses
+- Background processing for long operations
+
+## User Experience Vision
+
+### The Happy Path
+
+1. **Launch Application** → Welcome screen
+2. **Select Template** → "Basic Warehouse Analysis"
+3. **Upload Excel Files** → Drag & drop
+4. **Review Mapping** → Auto-detected, minimal adjustments
+5. **Import Data** → Progress bar, validation checks
+6. **View Dashboard** → Auto-generated overview
+7. **Run Analysis** → Click "ABC Classification"
+8. **Export Report** → One click, formatted Excel
+
+**Total Time**: < 10 minutes
+
+### Error Recovery
+
+- Clear error messages with context
+- "What went wrong" explanations
+- "How to fix" suggestions
+- One-click retry after fixes
+
+---
+
+**Version**: 1.0.0
+**Last Updated**: 2025-01-26
+**Status**: Draft