diff --git a/recommended-approach.md b/recommended-approach.md
new file mode 100644
index 0000000..620379c
--- /dev/null
+++ b/recommended-approach.md
@@ -0,0 +1,440 @@
+# Recommended Approach: 3-Phase Bootstrap Plan
+
+## Overview
+
+**Motto**: "Slow is smooth, smooth is fast"
+
+**Philosophy**: Bootstrap with minimal capital, use customer revenue to fund platform development, transition to recurring revenue model.
+
+---
+
+## Phase 1: Integration Consulting (Months 1-3)
+
+### Objective
+Generate cash flow, validate market demand, build network of relationships.
+
+### Timeline
+**Week 1-2**: Setup  
+**Week 3-4**: Outreach  
+**Week 5-8**: First pilot project  
+**Week 9-12**: Complete pilot, land second project
+
+### Services Offered
+- OT/IT integration consulting
+- PLC programming for data export
+- Edge gateway setup (on customer hardware)
+- MQTT/OPC UA configuration
+- Grafana dashboard deployment
+- Training and documentation
+
+### Revenue Model
+- **Hourly**: $150-200/hour
+- **Project**: $15k-40k (100-200 hours)
+- **Retainer**: $2k-5k/month (optional support)
+
+### Infrastructure
+**NONE** - Deploy on customer hardware:
+- Raspberry Pi 4 or old industrial PC (customer buys)
+- Ubuntu Server + Docker/LXC (free)
+- Mosquitto MQTT + InfluxDB + Grafana (free, open source)
+- Node-RED for PLC integration (free)
+
+### Capital Requirements
+**Total**: $500
+- LLC filing: $100-300
+- Basic website (Carrd/WordPress): $0-200
+- Business cards: $50
+- LinkedIn Premium: $0 (optional)
+
+### Activities
+
+#### Week 1-2: Foundation
+- [ ] Register LLC (1-2 hours, $100-300)
+- [ ] Set up website (4-6 hours, $0-200)
+  - Service description
+  - Case study section (empty initially)
+  - Contact form
+  - Bio highlighting OT + cloud expertise
+- [ ] Update LinkedIn profile (1 hour)
+  - "OT/IT Integration Consultant"
+  - Headline: "I help manufacturers bridge PLC data to dashboards"
+- [ ] Create contact list (20+ from controls engineering days)
+
+#### Week 3-4: Outreach
+- [ ] Email/call 20 contacts (5 per week)
+- [ ] Message template:
+  > "Hey [Name], after 13 years programming PLCs, I spent the last 4 years learning cloud engineering. Now I'm helping manufacturers get visibility into production data without expensive cloud platforms. Know anyone struggling with this?"
+- [ ] Target: 3-5 discovery calls
+- [ ] Join local ISA chapter (optional)
+
+#### Week 5-8: First Pilot
+- [ ] Land pilot project: $5k-10k (discounted)
+- [ ] Scope: Single production line, basic monitoring
+  - PLC data collection (Modbus/OPC UA)
+  - MQTT broker on Raspberry Pi
+  - Grafana dashboards (3-5 views)
+  - Alert configuration
+- [ ] Document EVERYTHING
+  - Before/after metrics
+  - Technical architecture
+  - Customer testimonial
+- [ ] Deliverables:
+  - Working edge gateway
+  - 3-5 Grafana dashboards
+  - Alert rules configured
+  - 2-4 hours of training
+  - Documentation for maintenance
+
+#### Week 9-12: Validate & Scale
+- [ ] Complete pilot project
+- [ ] Get written testimonial
+- [ ] Take photos/screenshots
+- [ ] Write case study
+- [ ] Publish to LinkedIn + website
+- [ ] Ask pilot customer for 2-3 referrals
+- [ ] Land second project at full price ($20k-40k)
+- [ ] Begin prospecting for Phase 2 recurring customers
+
+### Success Metrics (Phase 1)
+- [ ] 2-3 completed consulting projects
+- [ ] $20k-45k revenue generated
+- [ ] 5+ discovery calls completed
+- [ ] 1-2 case studies published
+- [ ] 10+ new LinkedIn connections
+- [ ] 3-5 warm prospects for Phase 2
+
+### Expected Outcomes
+- **Revenue**: $20k-45k in 90 days
+- **Profit**: ~$19k-43k (95% margin)
+- **Network**: 20+ new manufacturing contacts
+- **Validation**: Proven demand for services
+- **Foundation**: Customer base for Phase 2 transition
+
+---
+
+## Phase 2: Edge Monitoring Platform (Months 4-9)
+
+### Objective
+Transition from project-based to recurring revenue, deploy multi-tenant infrastructure.
+
+### Timeline
+**Month 4-6**: Build platform, convert 2-3 consulting clients  
+**Month 7-9**: Add 3-5 new monitoring customers, refine product
+
+### Services Offered
+**Consulting** (continued, reduced):
+- 1-2 projects/month for cash flow
+- Integration work for monitoring customers
+
+**New - Monitoring Platform**:
+- Edge monitoring as a service
+- Setup + monthly subscription
+- Remote management
+- Alert management
+- Data backup and retention
+
+### Revenue Model
+**Hybrid**:
+- Consulting: $10k-20k/month (declining)
+- Monitoring setup fees: $5k-10k per customer
+- Monitoring recurring: $0 → $10k/month (growing)
+
+### Infrastructure
+**GTHost Dedicated Server #1:**
+```
+Configuration:
+├── 8 cores, 32GB RAM, 1TB SSD
+├── Estimated cost: $100-150/month
+├── Capacity: 8-10 customers
+└── Revenue potential: $12k-30k/month
+```
+
+**Tech Stack:**
+- Ubuntu Server 24 LTS
+- LXC for container isolation (per customer)
+- MQTT broker per container
+- InfluxDB per container (or shared with DB per customer)
+- Grafana (shared with customer-specific dashboards)
+- Automated backup system
+- Monitoring and alerting (UptimeRobot, Prometheus)
+
+### Capital Requirements
+**Total**: ~$1,000
+- GTHost server: $100-150/month × 2 months
+- Development time: Nights/weekends (sweat equity)
+- Funded from Phase 1 consulting revenue
+
+### Activities
+
+#### Month 4-5: Platform Development
+- [ ] Provision GTHost server ($5/day trial first)
+- [ ] Build LXC container template
+  - MQTT broker (Mosquitto)
+  - InfluxDB instance
+  - Grafana user + dashboards
+  - Backup scripts
+- [ ] Create provisioning automation
+  - New customer onboarding script
+  - Container creation
+  - Credential generation
+  - DNS/subdomain setup
+- [ ] Build simple web portal
+  - Customer login
+  - Dashboard links
+  - Support ticket system
+  - Billing integration (Stripe)
+- [ ] Set up monitoring
+  - Server health monitoring
+  - Customer container monitoring
+  - Alert system for failures
+- [ ] Documentation
+  - Customer onboarding guide
+  - Internal runbook
+  - Support procedures
+
+#### Month 5-6: First Customers
+- [ ] Approach 2-3 existing consulting clients
+- [ ] Pitch: "I've built a monitoring platform. Want to be a pilot customer for $1k/month?"
+- [ ] Deploy for first 2 customers
+- [ ] Gather feedback, iterate
+- [ ] Test backup/restore procedures
+- [ ] Validate multi-tenant isolation
+- [ ] Begin marketing monitoring service
+
+#### Month 7-9: Growth & Refinement
+- [ ] Add 3-5 new monitoring customers
+- [ ] Continue 1-2 consulting projects/month (cash flow)
+- [ ] Refine platform based on feedback
+- [ ] Build case studies (monitoring-specific)
+- [ ] Automate more operations
+- [ ] Document standard procedures
+- [ ] Plan for second server (if needed)
+
+### Success Metrics (Phase 2)
+- [ ] GTHost server deployed and stable
+- [ ] 5-8 monitoring customers live
+- [ ] $10k-15k MRR achieved
+- [ ] Multi-tenant architecture proven
+- [ ] 99.5%+ uptime maintained
+- [ ] Customer churn <10%
+- [ ] 3+ new case studies published
+
+### Expected Outcomes
+- **Month 6 Revenue**: $17k ($15k consulting + $2k MRR)
+- **Month 9 Revenue**: $20k ($10k consulting + $10k MRR)
+- **Infrastructure Cost**: $150/month
+- **Profit Margin**: 85-90%
+- **Platform**: Production-ready, scalable
+- **Foundation**: Ready for Phase 3 premium features
+
+---
+
+## Phase 3: GPU-Powered AI Features (Months 10-18)
+
+### Objective
+Add premium AI-powered features for differentiation and margin expansion.
+
+### Timeline
+**Month 10-12**: Build AI infrastructure, launch premium tier  
+**Month 13-18**: Convert customers, add new premium customers
+
+### Services Offered
+**Basic Tier** (existing):
+- Edge monitoring
+- Dashboards and alerts
+- $1.5k-2k/month
+
+**Premium AI Tier** (new):
+- Predictive maintenance
+- Anomaly detection
+- Computer vision quality inspection
+- Advanced analytics
+- $4k-8k/month
+
+### Revenue Model
+**Fully Recurring**:
+- Basic customers: $1.5k × 8-10 = $12k-15k/month
+- Premium customers: $5k × 4-5 = $20k-25k/month
+- Consulting (occasional): $5k/month
+- **Total**: $37k-45k/month
+
+### Infrastructure
+**Server #1** (Basic tier):
+- 8 cores, 32GB, 1TB
+- 8-10 customers
+- $100-150/month
+
+**Server #2** (Premium AI tier):
+```
+Configuration:
+├── 16 cores, 64GB RAM
+├── NVIDIA Tesla P4 8GB GPU
+├── 2TB NVMe SSD
+├── Estimated cost: $350-450/month
+├── Capacity: 3-5 premium customers
+└── Revenue potential: $15k-40k/month
+```
+
+### Capital Requirements
+**Total**: ~$3,000
+- GPU server: $400/month × 3 months
+- ML model development time
+- Funded from Phase 2 profits
+
+### Activities
+
+#### Month 10-11: AI Infrastructure
+- [ ] Provision GTHost server with Tesla P4 GPU
+- [ ] Install ML frameworks
+  - TensorFlow or PyTorch
+  - Scikit-learn
+  - ONNX Runtime (optimized inference)
+- [ ] Build predictive maintenance models
+  - Anomaly detection (vibration, temperature)
+  - Failure prediction
+  - Remaining useful life (RUL)
+- [ ] Computer vision setup (if pursued)
+  - Object detection
+  - Quality inspection
+  - Defect classification
+- [ ] Test GPU inference performance
+- [ ] Create premium tier documentation
+
+#### Month 12: Premium Launch
+- [ ] Select 2 customers to upgrade (beta pricing)
+- [ ] Deploy AI features
+- [ ] Gather data for model training
+- [ ] Refine models based on customer data
+- [ ] Measure ROI for customers
+- [ ] Create case studies showing value
+
+#### Month 13-18: Scale Premium
+- [ ] Convert 2-3 more customers to premium
+- [ ] Market premium tier to prospects
+- [ ] Hire first contractor ($3k-5k/month)
+  - Customer support
+  - Deployment assistance
+  - Basic troubleshooting
+- [ ] Formalize processes
+- [ ] Build sales pipeline
+- [ ] Plan for additional servers if needed
+
+### Success Metrics (Phase 3)
+- [ ] GPU server operational
+- [ ] 4-6 premium tier customers
+- [ ] $35k+ MRR total
+- [ ] First employee/contractor hired
+- [ ] 95%+ customer satisfaction
+- [ ] 3+ AI-specific case studies
+- [ ] ROI demonstrated for customers
+
+### Expected Outcomes
+- **Month 18 Revenue**: $40k-50k/month
+- **Infrastructure Cost**: $500-600/month (2 servers)
+- **Contractor Cost**: $5k/month
+- **Net Profit**: $34k-44k/month
+- **Annual Run Rate**: $480k-600k
+- **Positioning**: Premium AI-powered industrial IoT platform
+
+---
+
+## Transition Strategy Between Phases
+
+### Phase 1 → Phase 2
+**Trigger**: When you have:
+- 2-3 completed consulting projects
+- $20k+ in the bank
+- 2-3 customers willing to pilot monitoring
+
+**Transition**:
+- Reduce consulting to 50% time
+- Invest 50% time in platform development
+- Use consulting cash to fund infrastructure
+- Convert consulting customers to monitoring
+
+### Phase 2 → Phase 3
+**Trigger**: When you have:
+- 5+ monitoring customers ($10k+ MRR)
+- Consistent $20k+/month total revenue
+- Stable platform with <5% churn
+- Cash reserves to fund GPU server for 3 months
+
+**Transition**:
+- Add GPU server (funded from Phase 2 profits)
+- Build AI features
+- Beta test with 2 willing customers
+- Launch premium tier
+- Hire first contractor to handle growth
+
+---
+
+## Risk Mitigation
+
+### Phase 1 Risks
+**Risk**: Can't find customers  
+**Mitigation**: 13 years of network, LinkedIn outreach, local ISA chapter
+
+**Risk**: Projects take too long  
+**Mitigation**: Scope small (40-80 hours max), resist feature creep
+
+**Risk**: Customers don't pay  
+**Mitigation**: 50% deposit upfront, milestone payments
+
+### Phase 2 Risks
+**Risk**: Platform technical failures  
+**Mitigation**: Start with 2-3 customers, test thoroughly, backup systems
+
+**Risk**: Can't transition consulting customers  
+**Mitigation**: Offer discounted pilot pricing, add value not available before
+
+**Risk**: Infrastructure costs escalate  
+**Mitigation**: Multi-tenant architecture, start with basic GTHost config
+
+### Phase 3 Risks
+**Risk**: AI features don't deliver value  
+**Mitigation**: Beta test with willing customers, measure ROI, iterate
+
+**Risk**: GPU costs too high  
+**Mitigation**: Start with 2-3 premium customers to fund server
+
+**Risk**: Can't hire good contractor  
+**Mitigation**: Use Upwork/Fiverr first, hire part-time before full-time
+
+---
+
+## Key Success Factors
+
+1. **Start Small**: $500 capital, basic consulting
+2. **Customer-Funded**: Use their money to build platform
+3. **Validate First**: Consulting proves demand before platform build
+4. **Recurring Focus**: Always transition to subscriptions when possible
+5. **Edge-First**: Maintains AWS non-compete compliance
+6. **Iterative**: Build → Test → Refine, don't try to be perfect
+7. **Network Effects**: Each customer = 2-3 referrals
+8. **Documentation**: Case studies critical for next phase
+
+---
+
+## Timeline Summary
+
+| Month | Phase | Focus | Revenue | Profit |
+|-------|-------|-------|---------|--------|
+| 1-2 | 1 | Setup, first project | $5k | $4.5k |
+| 3 | 1 | Second project | $20k | $19k |
+| 4-5 | 2 | Platform dev | $17k | $16k |
+| 6 | 2 | First monitoring customers | $20k | $19k |
+| 9 | 2 | Growing MRR | $20k | $19k |
+| 12 | 3 | Premium launch | $25k | $23k |
+| 18 | 3 | Full premium | $40k | $34k |
+
+**Cumulative 18 months**: ~$350k revenue, ~$300k profit
+
+---
+
+## Next Steps
+
+See `go-to-market.md` for detailed 90-day action plan.
+
+---
+
+*Last Updated: December 2025*