Data Platform Exec Summary
Source: docs/proposals/data-platform-exec-summary.md
# RGL8R: Regulatory Data Platform
**One-Pager for Executives**
---
## What We're Building
**RGL8R is a regulatory data platform with compliance applications on top.**
The applications (parcel audit, trade screening, landed cost) are the interface. The data is the product.
---
## The Problem
Global trade compliance requires accurate, up-to-date regulatory data:
| Data Type | Scale | Update Frequency |
|-----------|-------|------------------|
| Trade remedy orders (AD/CVD) | ~500 active globally | Monthly |
| HTS tariff schedules | ~18,000 items/country | Annually + interim |
| Carrier claim rules | 50+ carriers × rules | Quarterly |
| Sanctions lists | 10+ lists | Daily |
**Today:** This data is scattered across government websites, PDFs, and spreadsheets. Compliance teams manually track changes—or miss them.
**The cost of getting it wrong:**
- SIMA/AD duties: 20-200% of product value
- Missed carrier claims: $500-5,000 per shipment
- Sanctions violations: $250K+ fines, criminal liability
---
## Our Approach
### 1. Aggregate official sources
We ingest directly from government databases:
- CBSA, USITC, Federal Register, CBP rulings
- No dependency on expensive commercial data providers
### 2. AI-assisted curation
LLMs parse unstructured regulatory documents (Federal Register notices, scope rulings) and extract structured data. Humans review high-stakes items.
### 3. Versioned releases
Every screening decision pins to a specific data release (e.g., `CA-2026.02`). Full audit trail for compliance defense.
### 4. Coverage transparency
Every data point has a quality tier:
- **ALPHA:** Automated ingest only
- **BETA:** Automated + partial review
- **VERIFIED:** Full SME review + cross-source validation
Customers know exactly what they're getting.
---
## Competitive Moat
| Traditional Approach | RGL8R |
|---------------------|-------|
| License data from Descartes/Avalara ($50-500K/yr) | Own the data, control the roadmap |
| Data is a cost center | Data is the product |
| Generic global coverage | Deep coverage where customers need it |
| "Trust us" | Transparent provenance + audit trail |
**The moat deepens over time:** Every customer interaction improves data quality. Every edge case we resolve becomes a golden test case.
---
## Business Model
### Phase 1: Applications drive adoption
- Parcel audit (SHIP) and trade screening (TRADE) as entry points
- Data quality is a differentiator, not a product
### Phase 2: Data as a product
- API access for customers who want raw data
- Tiered pricing by coverage level and freshness SLA
### Phase 3: Platform for compliance
- Third-party apps built on our data layer
- Marketplace for specialized compliance tools
---
## Roadmap
| Phase | Timeline | Outcome |
|-------|----------|---------|
| **Foundation** | Q1 2026 | Database-driven data, provenance tracking |
| **North America** | Q2 2026 | Full CBSA SIMA + US AD/CVD + 12 carriers |
| **AI Curation** | Q3 2026 | Automated change detection + human review |
| **Global** | Q4+ 2026 | EU, UK, APAC markets |
---
## Why Now
1. **AI enables scale:** LLMs can parse regulatory documents that were previously manual-only
2. **Compliance burden increasing:** Tariff wars, sanctions, supply chain scrutiny
3. **SMB underserved:** Enterprise solutions are $100K+; SMBs have no good options
4. **Government sources digitizing:** More machine-readable data available than ever
---
## Ask
**Data is the foundation. Invest in it early.**
The applications are valuable, but they're views on top of the data. The data platform is the durable asset that compounds over time.
---
## Key Metrics to Track
| Metric | Phase 1 | Phase 2 | Phase 3 |
|--------|---------|---------|---------|
| Trade remedies covered | 8 | 450+ | 500+ |
| Data freshness SLA | Best effort | 7 days | 24 hours |
| Coverage tier (% VERIFIED) | 0% | 20% | 50% |
| Customer data quality NPS | Baseline | +10 | +20 |
---
*For technical details, see `docs/proposals/data-foundation-strategy.md`*