Scaling Mapping Operations

VPS MVP launched with 50 locations. The roadmap calls for 500 by year end. That's a 10x scaling challenge.

The Mapping Pipeline

For each location:

1. Site survey - Identify mapping area, access, timing
2. Capture - Collect imagery with calibrated equipment
3. Processing - Run SfM/MVS to create 3D map
4. QA - Verify accuracy, completeness, safety
5. Publishing - Make available to VPS queries
6. Maintenance - Monitor and update as needed

Each step has scaling bottlenecks.

Capture Scaling

Current state: In-house teams with specialized equipment

• Cost: $5,000+ per location
• Time: 1-2 days per location
• Capacity: 10 locations/month

Scaled approach: Partner network + consumer devices

• Mapping partners in multiple countries
• Lower-cost capture equipment
• Crowdsourced supplementary imagery
• Target: $500 per location, 100+ locations/month

Processing Scaling

Current state: Manual pipeline with expert oversight

• Processing time: 2-3 days per location
• Human review: 4 hours per location
• Failure rate: 20% require manual intervention

Scaled approach: Automated pipeline with exception handling

• Processing time: 8 hours (faster hardware, optimized code)
• Human review: 30 minutes (only exceptions)
• Failure rate: 5% require intervention (better preprocessing)

QA Scaling

Quality checks:

• Geometric accuracy: Does map match reality?
• Completeness: Are all important areas covered?
• Safety: Are there inappropriate elements?
• Consistency: Does it integrate with adjacent maps?

Current: Manual review of all outputs Scaled: Automated checks + sampling review

Automated checks catch 90% of issues. Human review for remaining risk.

Geographic Prioritization

Can't map everywhere immediately. Prioritization:

1. Developer demand: Where do partners want to build?
2. User density: Where are Quest users?
3. Strategic locations: Landmarks, venues, campuses
4. Capture efficiency: Clustered locations cheaper than scattered

Interactive prioritization dashboard for business decisions.

Operational Metrics

Tracking:

• Coverage velocity: Locations/month
• Cost efficiency: $/location
• Quality rate: % passing QA first time
• Time to publish: Capture to live
• Partner performance: Quality and speed by partner

Weekly operational reviews drive improvements.

The Chicken-and-Egg Problem

Developers want coverage before building. We want demand before expanding coverage.

Resolution:

• Seed coverage in strategic areas
• Committed partnerships in covered areas
• Demonstrate value to attract more developers
• Developer demand drives expansion

Breaking the cycle requires upfront investment in coverage.

Lessons from Month 1 of Scaling

• Partner training takes longer than expected
• Local regulations vary significantly by country
• Weather windows are real constraints
• Quality consistency across partners is challenging

Building robust processes before scale is essential.