10 KiB
Business Model Options
Overview
This document evaluates all business model options considered for the industrial IoT venture, with analysis of viability, capital requirements, and strategic fit.
Option 1: Integration Consulting ✅ RECOMMENDED (Phase 1)
Description
Provide OT/IT integration consulting services to manufacturers on a project or hourly basis.
Services Offered
- PLC programming for data export
- Edge gateway setup (on customer hardware)
- MQTT/OPC UA configuration
- Grafana dashboard deployment
- Cloud integration (AWS, Azure, GCP)
- Training for customer teams
Revenue Model
- Hourly: $150-200/hour
- Project-based: $15k-40k per project (100-200 hours)
- Small retainer: $2k-5k/month ongoing support
Economics
Capital Required: $500
- LLC filing: $100-300
- Basic website: $0-200
- Business cards/marketing: $50-100
Time to First Revenue: 30-60 days
Year 1 Potential: $100k-200k (5-10 projects)
Advantages
✅ Fast to revenue (30 days possible)
✅ Minimal capital required
✅ Validates market demand
✅ Builds network and case studies
✅ Customer-funded (use their hardware)
✅ No infrastructure costs
✅ Flexible schedule (works around full-time job)
Disadvantages
❌ Time-for-money (limited scaling)
❌ No recurring revenue
❌ Project-based lumpy cash flow
❌ Requires constant sales
Strategic Fit
Perfect for Phase 1: Generates cash, validates demand, builds relationships for future platform customers.
Option 2: Edge Monitoring Platform ✅ RECOMMENDED (Phase 2)
Description
Multi-tenant edge monitoring platform hosted on GTHost dedicated servers using LXC containers.
Services Offered
- MQTT broker per customer
- InfluxDB time-series database
- Grafana dashboards
- Alert management
- Remote monitoring and support
- Data backup and retention
Revenue Model
Service Tiers:
| Tier | Price/Month | Features |
|---|---|---|
| Basic | $1,500 | 10 sensors, 30 days retention |
| Pro | $2,000 | 50 sensors, 90 days retention, mobile app |
| Enterprise | $3,000+ | Unlimited sensors, 1 year retention, custom |
Setup Fee: $5k-10k one-time
Economics
Infrastructure Costs:
- GTHost server (8 cores, 32GB, 1TB): $100-150/month
- Capacity: 8-10 customers per server
- Revenue per server: $12k-30k/month
- Gross margin: 85-90%
Capital Required: $1,000
- First 2 months GTHost funded from consulting revenue
- Development time: nights/weekends
Break-even: 2-3 customers
Advantages
✅ Recurring revenue (sticky)
✅ Highly scalable (multi-tenant)
✅ High margins (90%+)
✅ Leverages LXC expertise from ZLH
✅ Edge-first (no AWS conflict)
✅ Can add customers to existing infrastructure
Disadvantages
❌ Requires infrastructure investment
❌ Technical complexity (multi-tenant)
❌ Support overhead
❌ Longer sales cycle than consulting
Strategic Fit
Perfect for Phase 2: Transitions consulting customers to recurring revenue, builds platform moat.
Option 3: Virtual PLC Management ⚠️ PHASE 2-3
Description
Deploy and manage soft PLCs (CODESYS) on customer hardware or GTHost servers.
Services Offered
- Soft PLC runtime deployment
- Remote management and monitoring
- PLC programming services
- Backup and disaster recovery
- Redundancy and failover
Revenue Model
- Setup: $10k-30k one-time
- Monthly management: $3k-8k/month per site
- Programming: $150-200/hour as needed
Economics
Infrastructure:
- Runs on customer hardware OR
- GTHost dedicated server: $150-300/month
- 1-3 customers per server (if hosted)
Capital Required: $2,000-5,000
- Development and testing
- Pilot deployment costs
Advantages
✅ Higher revenue per customer
✅ Sticky (critical infrastructure)
✅ Differentiated offering
✅ Leverages controls expertise
Disadvantages
❌ Higher liability (control systems)
❌ Longer sales cycle (6-12 months)
❌ Safety certification concerns
❌ More technical risk
Strategic Fit
Good for Phase 2-3: Once platform proven, add premium control tier for select customers.
Option 4: GPU-Powered AI/ML Services ✅ RECOMMENDED (Phase 3)
Description
Add AI-powered features using NVIDIA GPUs for predictive maintenance and computer vision.
Services Offered
- Predictive maintenance models
- Anomaly detection
- Computer vision quality inspection
- Advanced analytics dashboards
- Custom ML model development
Revenue Model
Premium Tiers:
| Tier | Price/Month | Features |
|---|---|---|
| AI Basic | $4,000 | Predictive maintenance, 20 assets |
| AI Pro | $6,000 | + Computer vision, 50 assets |
| AI Enterprise | $8,000+ | Custom models, unlimited assets |
Setup Fee: $10k-20k one-time (model development)
Economics
Infrastructure:
- GTHost server with Tesla P4: $350-450/month
- Capacity: 3-5 premium customers
- Revenue per server: $12k-40k/month
- Gross margin: 85%+
Capital Required: $5,000
- GPU server + development time
- Funded from Phase 2 profits
Advantages
✅ Premium pricing (high margins)
✅ Differentiated (competitors can't match)
✅ Defensible moat
✅ High customer value
✅ Multiple upsell paths
Disadvantages
❌ Requires GPU infrastructure
❌ More complex to build
❌ Longer sales cycle
❌ Need ML expertise (or partner)
Strategic Fit
Perfect for Phase 3: Once monitoring platform established, add AI tier for differentiation and margin expansion.
Option 5: Digital Twin / Virtual Commissioning ❌ NOT RECOMMENDED
Description
Build comprehensive digital twins and virtual commissioning environments for manufacturers.
Why NOT Recommended
❌ Extremely compute-intensive (high infrastructure costs)
❌ Long sales cycles ($50k-150k projects, 6-18 months)
❌ Competes with AWS Digital Twin services (non-compete risk)
❌ Requires team (can't do solo)
❌ Large enterprises only (SMB can't afford)
❌ Complex 3D/physics simulation (outside core expertise)
Alternative Approach
IF pursued (far future):
- Partner with simulation vendors
- Focus on PLC testing environments (simpler)
- Offer as premium add-on, not core service
- Wait until Phase 3+ with team in place
Option 6: Brownfield Retrofit Projects ✅ HYBRID OPPORTUNITY
Description
Add sensors and connectivity to existing equipment through project-based work.
Services Offered
- Wireless sensor installation (LoRaWAN, cellular)
- Edge gateway setup
- Data pipeline configuration
- Integration with monitoring platform
Revenue Model
- Project-based: $20k-50k per site
- Maintenance: $500-2k/month per site
Economics
Capital Required: ~$0 (use customer hardware)
Project Duration: 2-4 weeks
Projects per Year: 6-12 possible
Advantages
✅ Complements both consulting and platform
✅ Natural upsell to monitoring service
✅ Uses customer hardware (no capex)
✅ Project-based cash injection
Disadvantages
❌ Still time-for-money
❌ Requires travel to site
❌ Project-based (lumpy revenue)
Strategic Fit
Hybrid with Phases 1-2: Use as project revenue to fund platform development, convert to monitoring customers.
Rejected Options
MSP for General IT
Why NOT: Already saturated market, no OT differentiation, commoditized pricing
Cloud Platform Builder
Why NOT: Competes with AWS (non-compete violation), requires massive capital, too competitive
Hardware Sales
Why NOT: Low margins, logistics complexity, not our expertise
Traditional System Integrator
Why NOT: Project-only, doesn't leverage cloud expertise, mature competitive market
Recommended Business Model (Hybrid 3-Phase)
Phase 1 (Months 1-3): Consulting First
Revenue Model: Project + hourly
Target: $20k-45k in 90 days
Purpose: Cash flow, validation, network
Phase 2 (Months 4-9): Add Monitoring Platform
Revenue Model: Setup fees + recurring subscriptions
Target: $10k-15k MRR
Purpose: Recurring revenue, platform moat
Phase 3 (Months 10-18): Premium AI Features
Revenue Model: Premium subscriptions + consulting hybrid
Target: $35k+ MRR
Purpose: Differentiation, margin expansion
Revenue Mix Evolution
Month 3 (Phase 1)
- Consulting projects: 100%
- Recurring: 0%
- Total: $20k-45k/month
Month 9 (Phase 2)
- Consulting projects: 50%
- Recurring monitoring: 50%
- Total: $20k-30k/month
Month 18 (Phase 3)
- Consulting projects: 20%
- Recurring basic: 40%
- Recurring premium: 40%
- Total: $35k-50k/month
Critical Success Factors
What Makes This Work
- Start with Cash Flow: Consulting first = no runway stress
- Customer-Funded Platform: Their projects pay for infrastructure
- Recurring Revenue Focus: Transition to subscriptions ASAP
- Edge-First Always: Avoids AWS conflict, addresses data sovereignty
- Multi-Tenant Architecture: Scale efficiently with LXC expertise
- Premium Tier Later: Add GPU/AI only after base platform proven
What Would Make This Fail
- Starting with platform: No revenue, burning through capital
- Competing with AWS directly: Non-compete violation
- Trying to do everything: Scope creep, delayed launch
- Ignoring consulting revenue: Long path to profitability
- Building before validating: Wrong product-market fit
Financial Model Comparison
| Model | Year 1 Revenue | Gross Margin | Scalability | Capital Needed | Risk Level |
|---|---|---|---|---|---|
| Consulting Only | $100k-200k | 95% | Low | $500 | Low |
| Platform Only | $50k-120k | 90% | High | $10k+ | High |
| Hybrid (Recommended) | $150k-300k | 90% | High | $1,500 | Medium |
| Digital Twins | $0-50k | 40% | Medium | $50k+ | Very High |
| General MSP | $50k-150k | 60% | Medium | $5k | High |
Next Steps
See recommended-approach.md for detailed 3-phase execution plan.
Last Updated: December 2025