Rgl8r Product Blueprint V3.8
Source: docs/strategy/snapshots/rgl8r-product-blueprint-v3.8.md
# RGL8R: Complete Product Blueprint v3.7
**We enable compliant commerce. When product data is wrong, it costs money. We keep it right so brands can sell with confidence.**
> **Elevator Pitch:** See `brand/elevator-pitch.md` for the canonical pitch and variants.
> **Document Type:** Strategy & Vision Document
> **Purpose:** Defines product vision, market opportunity, and platform architecture
> **For Implementation Details:** See `specs/` folder for frozen engineering specifications
> **For Data Platform (internal, long-term):** See `rgl8r-platform/docs/proposals/data-foundation-strategy.md`
> and `strategy/data-platform-exec-summary.md`
---
**Last Updated:** February 2026
**Aligned With:** Dashboard UX Spec v2.6.9, Schema v1.9
---
## Time Horizon
This document has two time horizons:
- **2026 Execution Scope:** Canada-bound parcel audit (SHIP) and CBSA/SIMA compliance (TRADE) for US-based e-commerce brands.
- **Long-Term Platform Vision:** Full SHIP, TRADE, PRODUCT, FINANCE, GEOGRAPHIC modules across Canada, US, EU, UK, AU.
**How to read this document:**
- Sections 1-3, 7-8, and Phase 0-1 of the roadmap describe the 2026 execution scope.
- Sections 4-6, 9-11, and Phases 2-4 of the roadmap describe the long-term platform vision and are not 2026 commitments.
- The 2026 roadmap uses week-based planning; data-platform roadmaps use quarter-based planning (see data-platform exec summary).
---
## Why Now
Commerce compliance has become materially harder in the last 5 years:
- **Carrier complexity:** DIM-driven billing, fuel surcharge variability, and accessorial proliferation have made freight audit a full-time job
- **Cross-border enforcement:** CBSA modernization, SIMA expansion, and anti-dumping scrutiny have increased duty exposure
- **Catalog chaos:** Product data is increasingly incomplete, inconsistent, and scattered across systems
- **Regulatory velocity:** New measures, rate changes, and trade agreement updates happen faster than manual processes can track
Meanwhile, modern AI + rules systems now make attribute validation and evidence-backed decisioning practical at scale.
**RGL8R exists because compliance failures are now operationally frequent, financially material, and system-fixable.**
### The AI Inflection Point
> "AI doesn't fix a fragmented business. It exposes it."
AI tools are amplifiers. They amplify structure when structure exists. They amplify noise when noise exists. Two companies with identical datasets on paper get radically different value from the same AI tools—one has a digitized value chain, the other has a patchwork of spreadsheets, tribal knowledge, and systems that don't talk.
**RGL8R is the structure that makes AI work for compliance.** Without clean, validated SKU-level data, AI classification tools just amplify bad data faster. We're not competing with AI tools—we're the prerequisite for them working.
> "You cannot digitize 5% of the value chain and expect AI to work."
This is why partial solutions fail. You can't automate compliance on bad data. We make the data trustworthy first.
---
## 1. Product Overview
A modular compliance intelligence platform that enables compliant commerce. When product data is wrong, it costs money. We keep it right so brands can sell with confidence. RGL8R provides:
- multi-domain compliance automation (Ship, Trade, Finance, Product, Geographic)
- AI-powered attribute validation and HS classification
- carrier invoice audit, contract modeling, and freight bill reconciliation
- customs/trade compliance with CBSA, SIMA, landed cost calculation
- excise tax and alcohol/wine compliance automation
- global scalability (Canada → US → EU → UK → AU)
- configuration-driven rule engine that adapts to any regulatory requirement
- audit-proof logging and evidence generation for customs defense
Designed for Operations, Finance, Compliance, Catalog, IT, and Executive personas—each with role-appropriate depth but a unified compliance spine.
**The Core Insight:** Every shipping error is a compliance failure. Every customs delay is a classification problem. Every carrier overcharge is a contract enforcement gap. Every duty overpayment is a classification or origin error. RGL8R treats these as one unified regulatory challenge.
**Phase 1 Focus:** RGL8R is architected as a complete compliance operating system, with Phase 1 deliberately focused on parcel audit (SHIP) and CBSA/SIMA compliance (TRADE) to prove value before expanding. The platform vision is big; the 2026 execution is narrow and disciplined.
---
## 2. Validated Market Opportunity
### 2.1 The KnifeCenter Case Study
Real data from pilot customer analysis (January 2026):
| Metric | Value | Source |
|--------|-------|--------|
| Annual Savings Opportunity | **$37,800** | Q1-Q3 2025 data, annualized |
| Carrier Adjustment Bias | **87.3%** favor carrier (up from 69.6% in 2024) | UPS adjustment records |
| Accessorial Hit Rate | 46.5% of domestic UPS | RES/DAS/AHC charges |
| Weight Inflation | 86.9% of UPS shipments billed above actual | Bill vs actual weight |
| ROI at 15% Value Share | 6.7x | $37.8K Ă· $5,670 fee |
**Key Findings:**
- **Systematic, Not Accidental** — 87.3% of post-invoice adjustments favor the carrier. This isn't human error—it's a broken system getting worse (was 69.6% in 2024).
- **Accessorial Concentration** — Top 100 SKUs drive 30% of key accessorials (RES + DAS + AHC). Fix ~100 SKUs to capture 30% of the accessorial problem.
- **Weight Inflation is Pervasive** — 86.9% of UPS shipments billed above actual weight, averaging +2.34 lbs per shipment.
**Opportunity Breakdown:**
| Category | Est. Annual Impact |
|----------|-------------------|
| UPS Accessorial Validation | $17,500 (23% of UPS spend) |
| UPS Weight Inflation Prevention | $22,700 |
| FedEx Weight Inflation Prevention | $2,600 |
| **Total Opportunity** | **$37,800** |
**The Pitch:**
> "In our pilot data, 87% of carrier billing corrections favor the carrier—up from 70% the year before. That's not human error—that's a broken system getting worse. We found $37.8K/year in recoverable savings for a mid-market knife retailer. At 15% value share, that's 6.7x customer ROI."
### 2.2 The Wayfair Canada Opportunity
**Validated Pain (from call transcripts):**
| Period | SIMA Charges | Source |
|--------|--------------|--------|
| Nov 2025 (3 weeks) | **$935,000** | Dec 16 call |
| Dec 2025 (1 week) | **$695,000** | Dec 16 call |
| Monthly ongoing | **$300-450K** | Nov 12 call |
CBSA compliance analysis on 1,500 SKUs:
| Metric | Value |
|--------|-------|
| SKUs Analyzed | 1,500 |
| Red Severity (Critical) | 825 (55%) |
| Yellow Severity (Warning) | 188 (12.5%) |
| Green (Compliant) | 487 (32.5%) |
| Duty Recovery Opportunities | 12 SKUs |
**Value Story:** The Nov 2025 spike was caused by a rule bug that let 622 "true components" slip through in 9 days. Automated SIMA screening catches these errors before shipment.
**SIMA Outcomes (All 1,500 SKUs):**
| Outcome | Count | Description |
|---------|-------|-------------|
| AT_RISK | 37 | SKU matches active SIMA measure; action required |
| NEEDS_REVIEW | 0 | Missing attributes needed to determine exposure |
| CLEARED | 1,463 | No action required |
**CLEARED breakdown:**
| Reason | Count |
|--------|-------|
| NO_MEASURE_MATCH | 1,436 |
| OUT_OF_SCOPE_COUNTRY | 19 |
| OUT_OF_SCOPE_MATERIAL | 8 |
**Reporting note:** Customer-facing exports always include 1 row per SKU (row count matches input). The "in-scope subset" (64 SKUs where `hs_in_scope = true`) is a filtered view used internally for triage—it does not imply other SKUs were skipped or ignored.
**The $96K → $0 Story:**
Initial analysis flagged $96K/month in SIMA exposure. Manual validation revealed the classifications were actually correct—demonstrating RGL8R catches expensive mistakes others miss by validating actual product attributes, not just HS codes.
> "We handle messy reality. That's why we catch issues others miss—like the $96K SIMA exposure that turned out to be $0 when we validated actual product attributes."
### 2.3 Primary ICP for Phase 1
**Explicit focus for 2026:** US-based e-commerce brands and retailers shipping cross-border to Canada.
**Ideal Customer Profile:**
| Attribute | Requirement |
|-----------|-------------|
| Annual Parcel Spend | $1M+ |
| Canada Volume | Material (>10% of shipments or >$500K annual) |
| SKU Count | 1,000+ active SKUs |
| Current Pain | Manual carrier audits, customs delays, duty uncertainty |
| Tech Stack | Can export CSVs; bonus if Shopify/WooCommerce/ERP with API |
**Why This ICP First:**
- **Proven pain:** KnifeCenter and Wayfair both fit this profile
- **Canada complexity:** CBSA, SIMA, provincial taxes—enough regulatory surface to demonstrate platform value
- **Accessible decision-makers:** Ops/Finance leaders who feel the pain directly
- **Expansion path:** Same customers will want US domestic audit, then EU/UK
**Module applicability note:** SHIP is globally applicable immediately (domestic + cross-border parcel audit works anywhere). TRADE is Canada-first to prove audit-defense and SIMA value quickly. Our ICP is Canada-shipping brands because they experience both pains at once, but SHIP can land standalone in any parcel-heavy retailer.
**Explicitly NOT Phase 1 ICP:**
- Pure domestic US shippers (no cross-border pain yet)
- Alcohol/cannabis merchants (regulated goods require PRODUCT module)
- EU/UK-first importers (different regulatory stack)
- Marketplaces running their own customs infrastructure (Wayfair is the exception, not the rule)
---
## 3. Core UX Philosophy
Your north star:
> **Simple on the surface. Defensible under the hood. Confidence in every decision.**
**UX rules:**
- neutral, analytical tone
- confidence scores on every recommendation
- evidence blocks that enable audit defense
- collapsible depth—summary first, detail on demand
- no alert fatigue—surface issues only when confidence × impact threshold is met
- trust-first—show sources, methodology, regulatory references
- enterprise clarity—every action has an audit trail
- invisible complexity—AI does the work, user sees the answer
This differentiates you from legacy compliance tools that either overwhelm with data or provide black-box recommendations.
### 3.1 Explicit Scope Boundaries (What We're NOT Building in 2026)
Discipline signal for investors, customers, and ourselves. The vision is big. The 2026 execution is narrow.
**NOT in 2026:**
| Category | What We're Skipping | Why |
|----------|---------------------|-----|
| Geography | EU, UK, Australia expansion | Canada is complex enough; prove the model first |
| Modules | PRODUCT (alcohol, cannabis), FINANCE (payment restrictions), GEOGRAPHIC (sanctions) | Phase 2/3; need SHIP + TRADE revenue first |
| SHIP Scope | Full LTL freight audit, hazmat detection, multi-carrier optimization | Start with parcel audit for 1-2 carriers; expand after traction |
| TRADE Scope | Full duty recovery automation, global landed cost engine | Start with CBSA/SIMA risk + landed cost sanity checks |
| Export Controls | ITAR/EAR screening, ECCN classification | Phase 2; requires deep product classification first |
| Denied Party | OFAC, BIS, SEMA list screening | Phase 2; party data integration required |
| Infrastructure | SOC 2 Type II certification, 99.9% SLA commitments | Security hygiene yes; formal certification is 2027 |
| Integrations | Native ERP connectors, TMS/WMS bidirectional sync | CSV + basic API first; deep integrations after PMF |
| ML/AI | Anomaly detection, predictive classification, auto-learning | Rules + confidence scores first; ML layer after data volume |
**Architecture Note:** The corridor-agnostic architecture is built in Phase 1, even though only US→CA rules are populated. This ensures adding ITAR/EAR (origin rules) and denied party screening (party rules) in Phase 2 is configuration, not code change.
**What We ARE Building in 2026:**
| Capability | Scope | Exit Criteria |
|------------|-------|---------------|
| Parcel Invoice Audit | FedEx + UPS, core overcharge types (DIM, duplicate, service mismatch) | 5+ customers using monthly |
| CBSA/SIMA Review | HS confidence scoring, SIMA risk flags, basic landed cost checks | Wayfair pilot success |
| Evidence Generation | Exportable PDF/CSV with confidence scores and regulatory references | Customers use for audit defense |
| Data Ingestion | CSV upload, 1-2 carrier EDI formats, basic product schema | 80% of customer data auto-ingested |
| Pricing Operations | Baseline methodology, savings calculation, monthly reports | 3+ customers on value-based pricing |
**Trade-off Philosophy:**
> "We'd rather be excellent at parcel audit + CBSA compliance for Canada-bound e-commerce than mediocre at everything for everyone."
This means saying no to:
- "Can you add EU VAT?" → Not in 2026
- "Can you audit our LTL freight?" → Not in 2026 (unless it's a gate to a large deal)
- "Can you handle our alcohol shipments?" → Not in 2026
**Phase 1 MVP vs. Platform Backlog (Clarity):**
| Module | Phase 1 MVP (2026) | Post-PMF Backlog |
|--------|-------------------|------------------|
| SHIP | Parcel invoice audit for FedEx + UPS. Core overcharge types (DIM, duplicates, service mismatch, basic accessorial anomalies). CSV + 1 EDI format. | LTL audit, hazmat/DG detection, service optimization, full contract modeling automation, multi-carrier support. |
| TRADE | CBSA/SIMA screening for active measures (SSS, steel, aluminum). HS confidence workflow. Landed cost sanity checks for Canada. | Duty recovery automation, broader OGDs, multi-country rulesets (US/EU/UK), full landed cost engine, drawback automation. |
Everything in the "Post-PMF Backlog" column is described in this blueprint for architectural completeness, but is not a 2026 commitment.
---
## 4. Platform Architecture
### 4.1 Module Hierarchy
```
RGL8R (Regulatory Intelligence Platform)
│
├── PHASE 1 MODULES (Current Focus: KnifeCenter + Wayfair)
│ │
│ ├── SHIP Module (Carrier Compliance)
│ │ ├── Weight/Dimension Validation
│ │ ├── Freight Invoice Audit (Parcel + LTL)
│ │ ├── Contract Modeling & Rate Optimization
│ │ ├── Hazmat/DG Detection
│ │ ├── Carrier Rule Engine
│ │ └── Service Level Optimization
│ │
│ └── TRADE Module (Customs & Classification)
│ ├── HS Code Classification (AI + Rules)
│ ├── CBSA/Customs Compliance
│ ├── SIMA/Anti-Dumping Validation
│ ├── Landed Cost Engine
│ ├── Country-of-Origin Verification
│ ├── Trade Agreement Management (CUSMA, etc.)
│ ├── Intended Use Differentiation (Commercial vs Residential)
│ ├── Duty Recovery & Drawback
│ ├── Export Controls (ITAR/EAR) [Phase 2]
│ └── Denied Party Screening [Phase 2]
│
├── PHASE 2 MODULES (Post-PMF — target 2027)
│ │
│ ├── PRODUCT Module (Regulated Goods)
│ │ ├── Alcohol/Wine Compliance (State/Province)
│ │ ├── Cannabis Compliance
│ │ ├── Age Verification Requirements
│ │ ├── Licensing/Certification Tracking
│ │ ├── Platform Policy Compliance
│ │ └── Safety/Recall Monitoring
│ │
│ └── FINANCE Module (Payment & Tax)
│ ├── Payment Processor Restrictions
│ ├── High-Risk Merchant Detection
│ ├── Cross-Border Payment Rules
│ ├── Tax Nexus/VAT/GST Compliance
│ ├── Excise Tax Calculation
│ └── AML Screening
│
└── PHASE 3 MODULES (Scale — target 2028+)
│
└── GEOGRAPHIC Module (Regional Rules)
├── Sanctions Screening (OFAC, etc.)
├── Data Privacy (GDPR, CCPA)
├── Local Content Requirements
├── Advertising Restrictions
└── Marketplace Policy Mapping
```
**Phase 1 Focus:** SHIP + TRADE are the wedge. KnifeCenter validates freight audit; Wayfair validates CBSA/SIMA/landed cost. Everything else comes after these are proven and generating revenue.
### 4.2 Core Services Architecture
Ten interconnected services power the platform:
| Service | Function | Key Interface |
|---------|----------|---------------|
| Product Compliance Service (PCS) | Central source of SKU regulatory truth (HS codes, OGDs, origin, labeling, permits, excise) | `/sku/compliance` |
| Landed Cost Engine (LCE) | Internal duty/tax calculator using proprietary tariff & tax tables | `/pricing/internal` |
| External Calculator Adapters (ECA) | Normalized connectors to 3rd-party APIs (Avalara, Zonos, etc.) | `/pricing/external/:provider` |
| Validation & Audit Service (VAS) | Compares internal vs external outputs, scores variance, decides source of truth | `/pricing/validate` |
| Customs Documentation Service (CDS) | Generates CBSA/CRA documents (commercial invoice, B3/IID, permit attachments) | `/docs/*` |
| Import/Export Ledger (IEL) | Immutable record of all imports, exports, duty payments, and refunds | `/ledger/*` |
| Duty Recovery Service (DRS) | Handles drawback, duties-relief, and destruction refund logic | `/drawback/*` |
| Compliance Rules Engine (CRE) | Dynamic validation against OGD, SIMA, TDG, TRQ, labeling, and marking rules | `/rules/validate` |
| Freight Audit Service (FAS) | Parcel and LTL invoice audit, rate validation, accessorial verification | `/audit/freight` |
| Data Orchestration Layer (DOL) | Event bus routing between services | Event topics |
**Implementation Path:** Initial implementation uses an in-process module dispatcher and a single deployable app. As modules stabilize and boundaries are proven, we'll peel them into independent services behind the same interfaces. This avoids premature microservice complexity while maintaining clean separation of concerns.
**Event Topics:**
- `order.created`
- `order.shipped`
- `import.cleared`
- `export.confirmed`
- `variance.detected`
- `drawback.submitted`
- `classification.updated`
- `audit.completed`
### 4.3 Data Adapter & Attribute Validation Layer
**This is the moat.** All customer data flows through a normalization and attribute validation layer before it hits SHIP or TRADE modules. This is where we handle messy reality.
#### Phase 1 Scope: SIMA-Critical Attributes Only
In Phase 1, attribute validation is intentionally scoped to the minimal attributes required to decide exposure for the SIMA measures we support: **HS code, country of origin, and material**. Additional attributes (weight, dimensions, certifications, intended use, etc.) are parsed and stored but not yet validated unless required by an active compliance rule. This allows RGL8R to deliver defensible results quickly while preserving a clear expansion path.
#### Attribute Lifecycle: Parsed → Validated → Activated
| Stage | Definition | Example |
|-------|------------|---------|
| Parsed | Raw attribute extracted from customer data and stored | "Country of Origin: China" from Wayfair schema tag |
| Validated | Attribute normalized, verified against rules, stored with confidence | "CN" (ISO code), confidence 0.95, source: schema_tag |
| Activated | Validated attribute used in a compliance decision | CN origin triggers SIMA screening for SSS measure |
**Current state (v20):**
- **Parsed:** 2,591 unique attribute types from Wayfair data
- **Validated:** 3 attributes (HS code, country_of_origin, material) = 4,500 records for 1,500 SKUs
- **Activated:** Same 3 attributes power all SIMA risk decisions
This is a focused design for SIMA risk assessment, not a general attribute warehouse. The expansion path is clear: add attributes to validation as new compliance rules require them.
#### Why This Layer Exists
**The $96K→$0 SIMA story:** Initial analysis flagged $96K/month in anti-dumping exposure based on HS codes alone. When we validated actual product attributes (material composition, country of origin, intended use), the exposure disappeared—the classifications were correct, just missing context.
Without this layer, SIMA/ADD flags and landed-cost calculations would be based on incomplete or wrong data.
#### What It Does
| Function | Example |
|----------|---------|
| Aggregate multi-row attributes | Wayfair's schema tags spread across multiple rows per SKU → single unified record |
| Map customer-specific columns | Schema Tag / Attribute Title → attribute_name, Schema Tag / Attribute Response → attribute_value |
| Normalize carrier formats | FedEx EDI 210 vs UPS CSV vs DHL API → unified invoice schema |
| Validate critical attributes | Country of origin, material composition, intended use (commercial vs residential) |
| Store with confidence scores | Each validated attribute has source, timestamp, and confidence level |
| Handle missing data | Flag SKUs with incomplete attributes for manual review |
#### Customer-Specific Adapters
Each customer gets an adapter that handles their specific data format:
```typescript
// adapters/wayfair/index.ts
class WayfairAdapter {
async normalizeSchemaData(rows: WayfairSchemaRow[]): Promise<NormalizedSKU[]> {
// Aggregate multi-row schema tags into single SKU record
const grouped = groupBy(rows, 'SKU');
return Object.entries(grouped).map(([sku, attrs]) => ({
sku,
attributes: this.aggregateAttributes(attrs),
confidence: this.calculateConfidence(attrs)
}));
}
private aggregateAttributes(rows: WayfairSchemaRow[]): Record<string, string> {
// Map their column names to our standard fields
const mapping = {
'Country of Origin': 'origin_country',
'Primary Material': 'material',
'Intended Use': 'intended_use',
'Is upholstered?': 'is_upholstered'
};
return rows.reduce((acc, row) => {
const standardField = mapping[row.attribute_title];
if (standardField) {
acc[standardField] = row.attribute_value;
}
return acc;
}, {});
}
}
```
#### Attribute Validation Storage
Every validated attribute is stored with provenance:
```sql
CREATE TABLE attribute_validations (
id UUID PRIMARY KEY,
tenant_id UUID NOT NULL,
sku VARCHAR(100) NOT NULL,
attribute_name VARCHAR(100) NOT NULL,
original_value TEXT, -- What customer provided
validated_value TEXT NOT NULL, -- What we confirmed
source VARCHAR(100) NOT NULL, -- supplier_spec, commercial_invoice, etc.
confidence DECIMAL(3,2) NOT NULL,
validated_by VARCHAR(100), -- user or 'system'
validated_at TIMESTAMP DEFAULT NOW(),
UNIQUE(tenant_id, sku, attribute_name)
);
```
#### The Compounding Moat
Every validation we store makes the next one easier:
- **Pattern learning:** If 100 "Deervalley" sinks are stainless steel, the 101st probably is too
- **Supplier profiles:** Known suppliers have known material compositions
- **Category defaults:** Massage chairs from China are almost always upholstered seating
- **Cross-customer intelligence:** Anonymized patterns across customers improve confidence
Competitors starting fresh would need to rebuild this entire validation corpus from scratch.
### 4.4 Unified Customs Record
Shared data model across all services:
```json
{
"sku": "ABC-123",
"hs_code": "6109.10.00.00",
"origin_country": "US",
"declared_value": 85.00,
"weight_kg": 1.2,
"dimensions": { "l": 30, "w": 20, "h": 10, "unit": "cm" },
"duty_rate": 0.0,
"gst_rate": 0.05,
"pst_rate": 0.08,
"trade_agreement": "CUSMA",
"ogd_permits": [{"agency": "GAC", "permit_no": "TRQ-2025-001"}],
"labeling": {"bilingual": true, "origin_marked": true},
"excise": {"category": "wine", "abv": 13.5, "rate_per_litre": 0.687},
"dangerous_goods": {"un_number": "UN3481", "tdg_class": "9"},
"import_entry": {"cad_no": "24CAD123456", "release_date": "2025-10-10"},
"export_link": {"cers_no": "CERS-888888", "date": "2025-11-01"},
"audit_status": "verified",
"confidence_score": 0.94,
"evidence": []
}
```
---
## 5. Module Deep Dives
### 5.1 SHIP Module
The foundational module—where RGL8R proves value immediately.
#### Phase 1 Scope (2026)
What we're actually building this year:
- **Carriers:** FedEx and UPS parcel only
- **Audit types:** DIM/weight mismatches, duplicate charges, service level mismatches, obvious accessorial anomalies (residential vs commercial), basic rate vs contract checks
- **Inputs:** CSV upload, 1 FedEx EDI 210 format
- **Outputs:** Line-level variance flags with confidence scores, monthly savings report, invoice adjustment recommendations
*Everything else in this section describes the full SHIP vision, not the 2026 scope.*
#### Capabilities (Full Vision)
| Capability | What It Does | Value |
|------------|--------------|-------|
| Weight/DIM Validation | Detects discrepancies between catalog and carrier-billed weights | Prevents overcharges |
| Parcel Invoice Audit | Compares billed charges to contracted rates (FedEx, UPS, DHL, USPS) | Recovers 2-5% of spend |
| LTL Freight Audit | Validates linehaul, accessorials, fuel, cross-border fees | Recovers 3-8% of spend |
| Contract Modeling | Ingests carrier agreements, models discount structures | Ensures rates applied |
| DG/Hazmat Detection | Identifies dangerous goods from product attributes | Prevents delays, fines |
| Service Optimization | Recommends optimal service level by zone/urgency | Reduces costs |
#### LTL Freight Audit Process
For cross-border LTL into Canada:
1. **Gather Documents:** BOL, rate confirmation, commercial invoice, B3 form, broker invoice, POD
2. **Verify Linehaul:** Compare billed rate per CWT to contracted tariff
3. **Validate Accessorials:** Fuel surcharge, residential/liftgate, customs clearance, border crossing, detention
4. **Check Customs Costs:** Brokerage fees, duties/GST, disbursement fees, currency conversion
5. **Audit for Duplicates:** PROs billed by both carrier and broker
6. **Generate Variance Report:** By carrier, reason code, $ impact
#### Audit Types
| Audit | Logic | Confidence Threshold |
|-------|-------|---------------------|
| DIM Weight Mismatch | Carrier DIM > catalog DIM Ă— tolerance | 90% |
| Service Downgrade | Shipped economy, billed express | 95% |
| Duplicate Charge | Same tracking, same charge type, within 7 days | 98% |
| Accessorial Validation | Residential delivery on commercial address | 85% |
| Rate Card Mismatch | Billed rate > contracted rate for service/zone | 90% |
| Discount Leakage | Negotiated discount not applied | 95% |
| Fuel Surcharge Error | Wrong baseline index used | 90% |
### 5.2 TRADE Module
The differentiation layer—CBSA, SIMA, landed cost, duty recovery.
#### Phase 1 Scope (2026)
What we're actually building this year:
- **Geography:** Canada only
- **HS Classification:** Confidence scoring for existing HS codes; targeted reclassification for red/SIMA-sensitive SKUs
- **SIMA:** Screening for active measures (SSS, steel products, aluminum extrusions); risk flags based on HS, origin, and material
- **Landed Cost:** Sanity check on duty/tax vs existing calculator; variance detection, not full quoting engine
- **Outputs:** SIMA risk report, landed cost variance report, evidence blocks for audit defense
*Everything else in this section describes the full TRADE vision, not the 2026 scope.*
#### SIMA Outcome Taxonomy (User-Facing)
RGL8R returns exactly one SIMA outcome per SKU. The customer-facing report always includes 1 row per SKU submitted—row counts match input.
**User-Facing Statuses (3 states):**
| Status | Definition | User Action |
|--------|------------|-------------|
| AT_RISK | SKU matches an active SIMA measure based on HS + required attributes | Action required |
| NEEDS_REVIEW | SKU may be in scope, but required attributes are missing or below confidence threshold | Human confirmation required |
| CLEARED | No action required | None |
**CLEARED Reason Codes (preserved for auditability):**
| Reason Code | Definition |
|-------------|------------|
| NO_MEASURE_MATCH | HS does not match any active SIMA measures supported in Phase 1 |
| OUT_OF_SCOPE_COUNTRY | HS matches a measure category but origin country is not subject |
| OUT_OF_SCOPE_MATERIAL | HS matches a measure category but material does not meet measure criteria |
| EXEMPTION_APPLIED | SKU qualifies for a specific exemption (e.g., AE Phase 1) |
**The Trust Line:**
> "If we say CLEARED, we had enough data to clear it. If we don't have enough data, we say NEEDS_REVIEW."
**NEEDS_REVIEW Triggers (only when SKU is in the decision path):**
- Missing or low-confidence country of origin when HS is in scope (`hs_in_scope = true`)
- Missing or low-confidence material when the measure requires it
- Multiple candidate measures and cannot confidently select one
- HS was corrected and correction confidence is below threshold
Internally we preserve detailed reason codes for auditability, but customers see a simple 3-state outcome with optional drill-down.
#### Capabilities (Full Vision)
| Capability | What It Does | Value |
|------------|--------------|-------|
| HS Classification | AI-powered classification with confidence scores | Correct duty rates |
| Attribute Validation | Verifies origin, material, composition claims | Prevents misclassification |
| Intended Use Differentiation | Commercial vs residential classification using schema tags and product attributes | Correct duty treatment for upholstered seating, furniture |
| SIMA/ADD Screening | Screens against anti-dumping measures | Avoids 295% penalties |
| Landed Cost Engine | Calculates duties, taxes, fees by destination | Accurate pricing |
| CBSA Compliance | Validates against Canadian tariff schedule | Smooth clearance |
| Trade Agreement Mgmt | Applies CUSMA, CPTPP, etc. preferential rates | Duty savings |
| Duty Recovery | Drawback, duties-relief, destruction refunds | Recovers overpayments |
#### Landed Cost Calculation Logic
```python
def calculate_landed_cost(item, destination_province, origin_country, ship_method):
base_price = item.price
# 1. Calculate Duty
duty_rate = get_duty_rate(item.hs_code, origin_country)
if origin_country in ['US', 'MX'] and item.cusma_certified:
duty_rate = 0.0 # CUSMA eliminates duty
# Apply de minimis thresholds (CUSMA courier)
if base_price < 40 and ship_method == 'courier':
duty_rate = 0.0
tax_rate = 0.0
elif base_price < 150 and ship_method == 'courier':
duty_rate = 0.0 # No duty, but tax applies
duty = base_price * duty_rate
# 2. Calculate Tax (GST/HST/PST)
gst = 0.05 # Federal 5%
pst = get_provincial_pst(destination_province)
tax = (base_price + duty) * (gst + pst)
# 3. Add Shipping (excluded from duty base)
shipping = get_shipping_cost(ship_method, destination_province, item.weight)
# 4. Add Excise if applicable (alcohol, vaping, etc.)
excise = calculate_excise(item) if item.excise_category else 0
return {
"base_price": base_price,
"duty": duty,
"tax": tax,
"excise": excise,
"shipping": shipping,
"total_landed_cost": base_price + duty + tax + excise + shipping
}
```
#### Provincial Tax Rates (Canada)
| Province | GST | PST/HST | Total |
|----------|-----|---------|-------|
| Alberta | 5% | 0% | 5% |
| British Columbia | 5% | 7% | 12% |
| Ontario | 0% | 13% HST | 13% |
| Quebec | 5% | 9.975% QST | 14.975% |
| Nova Scotia | 0% | 15% HST | 15% |
#### SIMA/ADD Monitoring
Active measures tracked:
| Measure | Product Type | Duty Rate | Origin |
|---------|--------------|-----------|--------|
| Stainless Steel Sinks | Kitchen/bath sinks | 18.6%-35.5% | China |
| Carbon Steel Pipe | Welded carbon steel pipe | Variable | Multiple |
| Cold-Rolled Steel | Flat-rolled steel | Variable | Multiple |
| Corrosion-Resistant Steel | Coated steel | 10.9% | China |
| Aluminum Extrusions | Extruded aluminum | 18.4% | China |
#### SIMA Trigger Risk Assessment
Even without active measures, classification errors create risk:
```yaml
sima_trigger_risk:
high:
- severity: red
- is_china_origin: true
- is_target_category: true # e.g., upholstered seating
medium:
- severity: yellow
- is_target_category: true
low:
- severity: red
- is_target_category: false
none:
- severity: green
```
#### Duty Recovery (Drawback)
When goods are exported or destroyed after import:
1. Subscribe to `export.confirmed` and `destruction.logged` events
2. Match to original import (SKU + CAD number)
3. Calculate refundable duty (handles CUSMA "lesser-of" rule)
4. Generate K32/E15 forms
5. Submit to CRA/CBSA portal
6. Track recovery % and open claims
#### Export Controls (Phase 2)
**ITAR (International Traffic in Arms Regulations):**
- Screens products against US Munitions List (USML)
- Applies to ALL exports from US, regardless of destination
- State Department license required for controlled items
- Criminal penalties for violations
**EAR (Export Administration Regulations):**
- Screens products against Commerce Control List (ECCN)
- License requirements vary by destination country
- Commerce Department (BIS) administered
- Less severe than ITAR but broader scope
**RGL8R Approach:**
- Origin-based rules: fire when exporting FROM controlled jurisdiction
- Corridor-aware: EAR license requirements vary by US→[destination]
- Evidence generation for compliance documentation
```yaml
export_control_screening:
itar:
scope: origin
origin: US
action: BLOCK_PENDING_LICENSE
lists: [USML]
ear:
scope: corridor
origin: US
action: FLAG_FOR_REVIEW
license_required_by_destination: true
```
#### Denied Party Screening (Phase 2)
Screens all transaction parties against government prohibited party lists:
**US-Origin Transactions:**
- OFAC SDN (Specially Designated Nationals)
- OFAC Consolidated Sanctions
- BIS Denied Persons List
- BIS Entity List
- BIS Unverified List
- State Department Debarred List
**Canada-Origin/Destination:**
- SEMA (Special Economic Measures Act)
**Universal:**
- UN Security Council Sanctions
**Parties Screened:**
- Buyer/Purchaser
- Consignee
- End User
- Intermediate Consignee
- Freight Forwarder
- Financial institutions (for payment compliance)
**RGL8R Approach:**
- Fuzzy matching with confidence scores (names have variations)
- False positive management workflow
- Re-screening on list updates
- Audit trail for compliance documentation
```typescript
interface PartyScreenResult {
party: TransactionParty;
listsChecked: string[];
matches: PotentialMatch[];
outcome: 'CLEAR' | 'POTENTIAL_MATCH' | 'BLOCKED';
confidence: number;
requiresReview: boolean;
}
```
### 5.3 FINANCE Module
Payment compliance, tax, and excise.
#### Capabilities
| Capability | What It Does | Value |
|------------|--------------|-------|
| Payment Processor Rules | Flags products restricted by Stripe/PayPal | Prevents suspension |
| High-Risk Detection | Identifies sensitive categories | Proactive compliance |
| Cross-Border Payments | Currency and payment restrictions by country | Smooth checkout |
| Tax Nexus | State-by-state tax obligation tracking | Correct collection |
| Excise Tax Engine | Calculates federal/provincial excise (alcohol, vaping, cannabis) | Legal compliance |
#### Excise Tax Calculation (Alcohol)
```json
{
"alcohol": {
"abv": 13.5,
"volume_l": 0.75,
"product_type": "wine",
"excise_rate_per_litre": 0.687,
"provincial_markup_rate": 0.10,
"excise_licence": "EXC-CA-12345",
"import_permit": "LCBO-2025-789",
"warning_label": true
}
}
```
**Calculation:**
- Excise Tax = ABV Ă— volume Ă— federal rate
- Provincial Markup = provincial rate Ă— base value
- GST/HST = tax on (value + duty + excise + markup)
- Environmental fee/deposit (if applicable)
### 5.4 PRODUCT Module
Regulated goods compliance—alcohol, cannabis, firearms, age-restricted items.
#### Capabilities
| Capability | What It Does | Value |
|------------|--------------|-------|
| Alcohol Compliance | State/province shipping rules matrix | Legal compliance |
| License Verification | Tracks permits by state/product | Prevents illegal shipments |
| Age Verification | Integrates with verification providers | Legal protection |
| Cannabis Compliance | State-by-state rules (where legal) | Regulatory compliance |
| Platform Policies | Amazon, eBay, Shopify rules | Prevents delisting |
#### Alcohol Compliance Matrix
```yaml
states:
utah:
wine_direct_to_consumer: false
beer_shipping: false
spirits_shipping: false
license_required: "Utah DABC permit"
california:
wine_direct_to_consumer: true
beer_shipping: limited
spirits_shipping: false
license_required: "Type 02 or Type 17"
max_cases_per_month: 2
provinces:
ontario:
import_permit_required: "LCBO-XXXX"
excise_licence_required: true
direct_to_consumer: false # Must go through LCBO
british_columbia:
import_permit_required: "BC-LDB-XXXX"
direct_to_consumer: limited # Wine stores only
```
#### Carrier Restrictions for Alcohol
| Carrier | Alcohol Shipping | Requirements |
|---------|------------------|--------------|
| FedEx | Yes (licensed shippers) | Adult signature, age verification |
| UPS | Yes (licensed shippers) | Adult signature, special contract |
| USPS | No | Prohibited |
| DHL | Limited | Country-specific restrictions |
---
## 6. Confidence & Evidence System
The trust layer that enables audit defense.
### 6.1 Confidence Thresholds
| Category | Auto-Action | Suggest Only | Manual Required |
|----------|-------------|--------------|-----------------|
| HS Classification | ≥95% | 80-94% | <80% |
| SIMA Exposure | ≥90% | 75-89% | <75% |
| Freight Overcharge | ≥95% | 85-94% | <85% |
| Contract Violation | ≥90% | 80-89% | <80% |
| Attribute Validation | ≥85% | 70-84% | <70% |
| Landed Cost | ≥90% | 80-89% | <80% |
### 6.2 Evidence Block Structure
Every recommendation includes audit-ready evidence:
```yaml
evidence:
sku: "DRLL1526"
classification_error:
wayfair_hs: "6802.99.00" # Articles of stone
correct_hs: "7324.10.00" # Stainless steel sinks
primary_sources:
- type: "cbsa_tariff"
reference: "7324.10.00"
url: "cbsa-asfc.gc.ca/..."
- type: "sima_measure"
reference: "Stainless Steel Sinks ADD"
duty_rate: "18.6%-35.5%"
supporting_data:
- type: "product_attribute"
field: "material"
value: "304 Stainless Steel"
source: "supplier_specification"
- type: "product_attribute"
field: "country_of_origin"
value: "China"
source: "commercial_invoice"
methodology:
- "Compared product attributes against CBSA tariff schedule"
- "Cross-referenced SIMA measures database"
- "Validated material composition from supplier spec sheet"
confidence_factors:
attribute_completeness: 0.92
source_reliability: 0.95
rule_specificity: 0.88
overall: 0.94
cbsa_references:
- "CBSA Memorandum D10-0-1 Section 3.2"
- "CBSA Tariff Item 7324.10.00.00"
recommendation:
action: "Reclassify to 7324.10.00"
expected_impact: "$5,920/year duty exposure"
urgency: "high"
```
### 6.3 Audit Defense Package Generation
One-click export for CBSA inquiries:
- Classification history with timestamps
- Evidence chain for each decision
- Regulatory references (CBSA memos, tariff items)
- Attribute validation trail
- PDF format suitable for regulatory submission
---
## 7. Proportional Value-Based Pricing
### 7.1 The Core Philosophy
**Charge 15% of demonstrated value. Uncapped by default.**
| Customer | Saves | Pays (15%) | Their ROI |
|----------|-------|------------|-----------|
| Small retailer | $10K/mo | $1,500/mo | 6.7x |
| KnifeCenter | $18.6K/mo | $2,788/mo | 6.7x |
| Wayfair | $200K/mo | $30K/mo | 6.7x |
| Fortune 500 | $2M/mo | $300K/mo | 6.7x |
Everyone gets the same 6.7x ROI. No one subsidizes anyone else.
**Enterprise note:** Pricing is uncapped by default. For very large accounts, we can offer structured commitments (minimums, quarterly true-ups) in exchange for longer terms and defined scope, without changing the core value-capture model.
### 7.2 Pilot vs Standard Pricing
**Important context on case study numbers:**
Early pilot pricing with KnifeCenter was $500/month, which delivered a 37x ROI and proved the concept. This aggressive pilot pricing was intentional—it removed all friction and let us validate the value proposition with real data.
Going forward, our standard model is **15% of recovered value**, which delivers ~6.7x ROI for customers. For KnifeCenter, that means moving from $500/month to ~$2,788/month—still an exceptional ROI, and now a sustainable business model.
**The pilot-to-standard conversation:**
> "We started you at $500/month to prove this works. The data shows we're saving you $223K/year. At our standard 15% model, that's $2,788/month—you still keep 85% of the savings, and we can invest in building out the features you've asked for."
### 7.3 Why 15% (Not 50% or 100%)
| Value Capture | Customer ROI | Outcome |
|---------------|--------------|---------|
| 100% | 1x (break-even) | No sale—why bother? |
| 50% | 2x | Too risky for CFO approval |
| 25% | 4x | Decent, but shopping risk |
| 15% | 6.7x | Instant yes, no shopping |
| 10% | 10x | Leaving money on table |
**Strategic reasons for 15%:**
- **Sales Velocity:** Under $3K/month = manager approval (not C-suite)
- **Customer Success:** They become evangelists at 6.7x ROI
- **Competitive Moat:** Too much value to switch providers
- **Expansion Revenue:** Room to add modules and grow
- **Market Penetration:** Lower friction = faster growth = higher valuation
### 7.4 Enterprise Pricing Structure
For enterprise accounts (Wayfair-scale), pricing follows the same value-based model with additional structure:
**15% of demonstrated savings, with:**
- Minimum monthly commitment (ensures engagement and support coverage)
- Quarterly true-up based on actual savings realized
- Module-specific pricing when using multiple modules
**Positioning vs. commodity classifiers:**
> "Zonos sells HS predictions for pennies. We sell audit relief, valuation control, and CBSA-proof evidence. They're a component; we're the solution."
### 7.5 Module Stacking Economics
Customers naturally expand from 1 → 2.8 modules average:
**Single Module Customer:**
- Uses SHIP only
- Saves $50K/month → Pays $7,500/month
**Multi-Module Customer:**
- SHIP: Saves $50K → Pays $7,500
- TRADE: Saves $80K → Pays $12,000
- PRODUCT: Saves $30K → Pays $4,500
- **Total:** Saves $160K/month → Pays $24,000/month
Still 6.7x ROI across all modules!
### 7.6 Pricing Configuration
```yaml
# modules/pricing/rules.yaml
pricing:
value_capture_rate: 0.15 # 15% of savings
ship:
freight_audit:
value_capture: 0.15
min_monthly: 500
catalog_enrichment:
value_capture: 0.15
trade:
hs_classification:
value_capture: 0.15
duty_optimization:
value_capture: 0.15
sima_prevention:
value_capture: 0.15
product:
alcohol_compliance:
value_capture: 0.15
includes: ["state_matrix", "age_verification", "licensing"]
finance:
excise_calculation:
value_capture: 0.15
payment_compliance:
value_capture: 0.15
```
### 7.7 Pricing Operationalization: Baseline Methodology & Dispute Resolution
The 15% model is conceptually clean. Operationalizing it requires clear methodology.
#### Baseline Definition
**For Freight Audit (SHIP):**
- **Baseline** = carrier invoices as billed
- **Savings** = (billed amount) - (correct amount based on actual weight/DIM, contracted rates, service used)
- **Methodology:** Compare invoice line items against validated shipment data and contract terms
**For CBSA/SIMA (TRADE):**
- **Baseline** = duty/tax as calculated by current process (checkout calculator, broker estimates)
- **Savings** = (baseline duty) - (correct duty based on validated HS codes, origin, trade agreements)
- **Additional value:** audit cost avoidance, SIMA penalty prevention (harder to quantify, use industry benchmarks)
#### Savings Calculation Timing
RGL8R tracks three distinct value categories, each with different billing triggers:
| Category | Definition | Billing Trigger | Example |
|----------|------------|-----------------|---------|
| Recovered | Cash returned via carrier credit or refund | When credit is issued or invoice adjusted | Duplicate charge refund, late delivery credit |
| Avoided | Savings realized by preventing overcharge before payment | When invoice posts at correct (lower) amount | DIM correction applied before invoice approval |
| Risk Prevented | Exposure eliminated that would have resulted in penalties or duties | Proxy billing (per SKU flagged, per audit package, % of exposure value) | SIMA exposure identified and reclassified before import |
**Why this matters for enterprise pricing:**
- **Recovered** is easiest to bill—clear counterfactual, customer sees the credit
- **Avoided** requires baseline agreement—customer must accept "what would have been billed" methodology
- **Risk Prevented** is hardest—no cash changes hands, value is counterfactual; bill via proxy metrics
**Practical application:**
| Scenario | Category | Billing Approach |
|----------|----------|------------------|
| Carrier issues $500 credit for duplicate charge | Recovered | Bill 15% ($75) on credit confirmation |
| DIM correction reduces invoice from $1,000 to $800 | Avoided | Bill 15% of $200 ($30) on invoice approval |
| SIMA reclassification prevents $50K exposure | Risk Prevented | Bill flat fee per SKU flagged OR % of exposure prevented |
| Audit cost reduction (fewer broker hours) | Risk Prevented | Bill based on audit hours saved Ă— internal rate |
KnifeCenter demonstrates Recovered + Avoided. Wayfair demonstrates Risk Prevented. Together they justify why RGL8R is a compliance decision engine, not just a freight audit tool.
#### Dispute Resolution
**Principle:** Customer always sees the math. No black boxes.
**Process:**
1. Monthly savings report shows: baseline, actual, variance, savings claimed, fee calculation
2. Customer has 10 business days to dispute specific line items
3. Disputed items are reviewed jointly; if customer is right, credit issued
4. Undisputed savings are billed
**Guardrails for Enterprise:**
- Minimum monthly commitment (e.g., $5K/month for Wayfair-scale) ensures engagement
- Soft cap subject to renewal negotiation (e.g., "15% of savings up to $50K/month; above that, renegotiate")
- Quarterly true-up: if actual savings exceed/underperform estimate, adjust next quarter
#### The KnifeCenter Transition
Specific plan for moving from pilot to standard pricing:
| Phase | Timeline | Price | Rationale |
|-------|----------|-------|-----------|
| Pilot | Months 1-3 | $500/mo | Prove value, no friction |
| Transition | Month 4 | $1,500/mo | 3x increase, still 15x ROI |
| Standard | Month 6+ | $2,788/mo (15% of $18.6K) | Full value-based pricing |
**Conversation script:**
> "We've proven $223K in annual savings. At $500/month, that's a 37x ROI—which was intentional for the pilot. Our standard model is 15% of savings. For you, that's $2,788/month—still a 6.7x ROI, and it lets us invest in the features you've asked for like [specific feature]. We can phase in over 2 months if helpful."
---
## 8. Sales Motion: The Audit Hook
### 8.1 The Proven Playbook
**Free Audit → Show Losses → Pilot → Close → Expand**
**Step 1: Free Shipping Data Audit**
- Customer uploads CSV of shipments or invoices
- System analyzes for errors, overcharges, misclassifications
- Benchmark: KnifeCenter's 91.8% error rate
**Step 2: Shocking Loss Report**
- Generate personalized ROI showing:
- Total annual losses to bad data
- Error rate vs. industry benchmark
- Top 10 highest-impact issues
- Projected savings at 15% fee
**Step 3: 30-Day Pilot**
- "We'll find 10x our fee or money back"
- Prove value on real data
- Build trust and integration
**Step 4: Close**
- Upgrade to proportional pricing
- Start at one module
- Natural expansion path
**Step 5: Expand**
- Use their data to show module value:
- "You ship to 37 countries—Trade Module would save $X"
- "Your 2Day/Ground mix is off—Optimization would save $Y"
- "You ship alcohol—Product Module prevents $Z in penalties"
### 8.2 The CFO Close
Frame pricing against alternatives:
- "Cheaper than 1 shipping clerk" ($4K/month = $48K/year)
- "Prevents one customs penalty per month"
- "Costs less than your monthly DIM surcharges"
### 8.3 AI Agent Discovery Channel
AI assistants (Claude, ChatGPT, Copilot, custom agents) increasingly help users solve operational problems. When a user asks an AI "how do I audit my FedEx invoices?" or "check my catalog for SIMA exposure," RGL8R should be in the consideration set.
**Why This Matters:**
- **Zero-CAC acquisition** — AI agent recommends us based on capability match
- **Trust transfer** — user trusts the agent's recommendation
- **Natural expansion** — agent can suggest additional modules based on user's data
- **Automation-native** — users building automated workflows discover us programmatically
**Discovery Enablers:**
| Asset | Purpose | Location |
|-------|---------|----------|
| `llms.txt` | Capability description for AI crawlers | rgl8r.com/llms.txt |
| OpenAPI Spec | Semantic endpoint descriptions with use-case context | api.rgl8r.com/docs |
| MCP Server | Direct tool access for AI agents | MCP registry + npm |
| CLI | Scripted automation, CI/CD integration | npm, GitHub |
**Positioning for AI Discovery:**
The key is **semantic clarity** — AI agents need to understand *what problem we solve*, not just *what endpoints exist*.
> **Bad:** `POST /api/v1/sima/screen` — "Screens SKUs against SIMA measures"
>
> **Good:** `POST /api/v1/sima/screen` — "Check if products you're importing to Canada face anti-dumping duties (SIMA). Upload your catalog; get back which SKUs are at risk of 18-295% additional duties and which are cleared."
**Integration with Sales Motion:**
AI agent discovery feeds into the standard sales funnel:
1. **Agent recommends RGL8R** for user's compliance problem
2. **User runs free audit** via CLI or API (same as web flow)
3. **Shocking Loss Report** generated automatically
4. **Human sales touch** for pilot conversion (or self-serve for SMB)
---
## 9. Global Scalability
### 9.1 Architecture Readiness
| Element | Status | Notes |
|---------|--------|-------|
| Microservice model | âś… Ready | Every domain independent |
| Unified customs schema | âś… Ready | Corridor as first-class field |
| Rules engine design | ✅ Ready | Corridor-aware rule inheritance (universal → origin → destination → corridor → party → product → finance → carrier) |
| Event-driven orchestration | âś… Ready | Branch by corridor, not just country |
| Export controls | 🔜 Phase 2 | ITAR/EAR screening for US origin |
| Denied party screening | 🔜 Phase 2 | OFAC, BIS, SEMA list matching |
**Corridor-Agnostic Design Principle:**
The platform treats corridor (origin → destination) as a first-class parameter, not a hardcoded assumption. This means:
- **Same SKU, multiple corridors:** A product can be shipped US→CA and US→UK with different rule evaluations
- **Origin rules + destination rules:** ITAR (US origin) and SIMA (CA destination) both fire on a US→CA shipment
- **New corridor = new rules, not new code:** Adding UK support means populating `/rulesets/destination/uk/` and `/rulesets/corridor/us_uk/`
See `strategy/self-extending-architecture-v1.md` for the full rule inheritance model.
### 9.2 Country Expansion Roadmap
| Phase | Countries | Key Requirements |
|-------|-----------|------------------|
| Phase 1 | Canada (current) | CBSA, SIMA, GST/HST/PST |
| Phase 2 | United States | USHTS, FDA, TTB, state sales tax |
| Phase 3 | European Union | TARIC, VAT OSS, CE marking |
| Phase 4 | United Kingdom | UK Global Tariff, HMRC, EMCS |
| Phase 5 | Australia | ICS, ATO, GST, WET |
### 9.3 Multi-Country Rule Structure (Corridor-Aware)
Rules are organized by scope, not just destination. The architecture supports any origin → any destination:
```
/rulesets/
├── universal/ # Apply everywhere
│ ├── physics.yaml # Weight/dims validation
│ ├── un_hazmat.yaml # UN dangerous goods codes
│ └── hs_harmonized.yaml # First 6 digits of HS
│
├── origin/ # Apply to exports FROM country
│ ├── us/
│ │ ├── itar.yaml # Defense articles (USML)
│ │ ├── ear.yaml # Dual-use (ECCN)
│ │ └── denied_parties.yaml # OFAC, BIS lists
│ └── ca/
│ └── sema.yaml # Canadian sanctions
│
├── destination/ # Apply to imports TO country
│ ├── ca/
│ │ ├── tariff.yaml
│ │ ├── gst_hst_pst.yaml
│ │ ├── sima.yaml
│ │ └── ogd.yaml
│ ├── us/
│ │ ├── tariff.yaml
│ │ ├── section_301.yaml
│ │ ├── fda.yaml
│ │ └── ttb.yaml
│ ├── eu/
│ │ ├── taric.yaml
│ │ ├── vat.yaml
│ │ └── excise.yaml
│ └── uk/
│ ├── tariff.yaml
│ └── hmrc.yaml
│
├── corridor/ # Apply to specific origin→destination
│ ├── us_ca/
│ │ ├── cusma.yaml # Trade agreement
│ │ └── de_minimis.yaml # Threshold rules
│ ├── cn_us/
│ │ └── section_301_tariffs.yaml
│ └── us_uk/
│ └── post_brexit.yaml
│
├── product/ # Apply to product categories
│ ├── alcohol.yaml
│ ├── cannabis.yaml
│ ├── perfume_fragrance.yaml
│ └── firearms.yaml
│
├── finance/ # Apply to payment/transaction
│ ├── payment_processors.yaml
│ ├── aml_thresholds.yaml
│ └── tax_nexus.yaml
│
└── carrier/ # Apply to shipping method
├── fedex.yaml
├── ups.yaml
└── dhl.yaml
```
**Rule Evaluation:** Engine loads applicable rulesets based on transaction attributes:
1. Always load `universal/`
2. Load `origin/{origin_country}/` based on shipper location
3. Load `destination/{destination_country}/` based on delivery location
4. Load `corridor/{origin}_{destination}/` if exists
5. Load `product/{category}/` based on product attributes
6. Load `finance/` rules based on transaction attributes
7. Load `carrier/{carrier}/` based on shipping method
**This is the corridor-agnostic architecture.** Adding a new corridor (e.g., US→UK) means populating the relevant ruleset folders, not changing code.
---
## 10. User Stories by Department
### 10.1 Finance Department
**Invoice Auditing & Validation**
As a Finance Manager, I want the system to automatically ingest carrier invoices (EDI 210s, PDFs, CSVs) and validate charges against contracted rates, so that I can identify overcharges before approving payment.
*Acceptance Criteria:*
- System flags any rate higher than contract
- Unexpected accessorial fees highlighted
- Tolerance threshold configurable (e.g., auto-approve within 1%)
- Dashboard shows pending review vs. auto-approved
**Duty Reconciliation**
As a Finance Analyst, I want to reconcile duties/taxes collected at checkout against amounts paid at customs, so that I can identify variances and improve pricing accuracy.
*Acceptance Criteria:*
- Per-shipment comparison: collected vs. paid
- Variance flagging with root cause
- Monthly reconciliation report
- Adjustment recommendations
### 10.2 Logistics Department
**Shipment Tracking & Visibility**
As a Logistics Manager, I want consolidated tracking across all carriers (parcel + LTL), so that I can monitor on-time performance and intervene on delays.
*Acceptance Criteria:*
- Single view across FedEx, UPS, DHL, LTL carriers
- Alerts for exceptions (late, damage, held at customs)
- Filter by status, carrier, date, customer
**Service Performance Auditing**
As a Transportation Analyst, I want carrier performance audited against SLA, so that I can claim refunds and address service issues.
*Acceptance Criteria:*
- Automatic late delivery detection
- Refund eligibility flagging
- Performance scorecard by carrier/lane
### 10.3 Compliance Department
**HS Classification Review**
As a Compliance Manager, I want to review AI-suggested HS classifications with confidence scores, so that I can approve or correct before customs declarations.
*Acceptance Criteria:*
- Queue of classifications awaiting review (Yellow/Red tier)
- Evidence block showing AI reasoning
- One-click approve or override with reason
**SIMA/ADD Monitoring**
As a Trade Compliance Analyst, I want automatic screening against SIMA measures, so that I identify exposure before goods ship.
*Acceptance Criteria:*
- Real-time screening against all active measures
- Exposure report with duty impact
- Alert when new products match SIMA criteria
- Customer-facing report includes every SKU submitted (1 row per SKU) with outcome = AT_RISK, NEEDS_REVIEW, or CLEARED, plus a reason code
**Audit Defense Package**
As a Compliance Director, I want to generate audit defense packages, so that I can respond to CBSA inquiries with documented evidence.
*Acceptance Criteria:*
- One-click export: classification history, evidence, sources
- Regulatory references included
- PDF format for regulatory submission
### 10.4 IT Department
**Platform Agnostic Deployment**
As an IT Manager, I want containerized deployment, so that I can run on any cloud or on-premises.
*Acceptance Criteria:*
- Docker containers for all services
- Kubernetes/Helm charts
- No proprietary cloud dependencies
- AWS, GCP, Azure, on-prem documentation
**API-First Integration**
As an Integration Engineer, I want every function accessible via API, so that I can connect ERP, WMS, TMS.
*Acceptance Criteria:*
- OpenAPI/Swagger documentation
- Webhook support
- API versioning
- OAuth2 + API key authentication
### 10.5 AI Agents & Automation Partners
**CLI Integration**
As a DevOps Engineer, I want a CLI tool that integrates RGL8R into our data pipelines, so that compliance checks run automatically on new catalog uploads or invoice batches.
*Acceptance Criteria:*
- CLI installable via npm/brew
- Commands for SHIP audit and TRADE screening
- JSON output for pipeline parsing
- Documented exit codes (0=success, 1=findings, 2=error)
- Integration keys with tenant binding (no shared secrets)
- PII sanitization with fail-closed policy
**MCP Server Access**
As an AI Agent Developer, I want RGL8R available as an MCP server, so that AI assistants can invoke compliance checks as tool calls within user conversations.
*Acceptance Criteria:*
- MCP server published to registry
- Tools for: `screen_catalog`, `audit_invoices`, `check_hs_code`, `estimate_landed_cost`
- Semantic tool descriptions optimized for agent understanding
- Structured output that agents can interpret and act on
- Rate limiting and auth via integration keys
**AI Agent Discovery**
As a user asking an AI assistant for help, I want the assistant to know about RGL8R when my question involves carrier invoice auditing or customs compliance, so that I discover the right tool for my problem.
*Acceptance Criteria:*
- `llms.txt` published at rgl8r.com describing capabilities
- OpenAPI spec includes semantic descriptions (problem → solution framing)
- Documentation structured for AI comprehension (clear headings, examples)
- CLI `--help` output optimized for agent parsing
---
## 11. Pain Point → Solution Mapping
| Pain Point | Solution | Phase | Impact |
|------------|----------|-------|--------|
| Disparate data formats | Multi-format ingestion (API, CSV, EDI, PDF/OCR) | 1-2 | 95% auto-ingestion |
| Billing errors & overpayments | Automated audit rules + tolerance-based approval | 1-2 | 3-5% spend recovered |
| Siloed processes | Unified platform + ERP/TMS/WMS integration | 2 | Minutes vs. hours to investigate |
| No visibility/analytics | Dashboards + BI integration | 2 | Real-time spend visibility |
| Classification errors | AI classification + attribute validation | 1 | 95%+ accuracy |
| SIMA/ADD exposure | Real-time screening + evidence generation | 1 | $0 unexpected duties |
| Customs audit risk | Evidence blocks + audit defense packages | 1 | Pass audits with documentation |
| Difficult partner onboarding | Modular adapters + multi-tenant | 2-3 | <1 week new carrier |
| Cross-border landed cost | Landed Cost Engine + dynamic pricing | 2 | ≤1% variance |
| Duty overpayments | Duty Recovery Service + K32/E15 automation | 3 | Recover historical overpayments |
| ITAR/EAR export violations | Export control screening (origin-based rules) | 2 | Avoid criminal penalties |
| Denied party transactions | Party screening against OFAC, BIS, SEMA lists | 2 | Block prohibited transactions |
| Multi-corridor complexity | Corridor-agnostic rule engine | 1+ | Same SKU, any origin→destination |
---
## 12. Implementation Roadmap
### Phase 0: Wayfair Pilot (Now → Dec 15, 2024)
- CBSA compliance analysis
- SIMA exposure validation
- Attribute validation proof
- Evidence block generation
**Exit Criteria:** Demonstrated $400K+ annual value
### Phase 1: Foundation (Weeks 1-8)
| Week | Deliverable |
|------|-------------|
| 1-2 | Database schema, API scaffold, auth (JWT + tenant), module registration |
| 3-4 | SHIP Module MVP: FedEx/UPS API adapters, invoice ingestion, basic audits |
| 5-6 | EDI 210 parser, message queue, freight audit for KnifeCenter |
| 7-8 | TRADE Module MVP: HS classification, CBSA tariff lookup, SIMA screening |
**Exit Criteria:**
- KnifeCenter freight audit live
- Wayfair classification pilot complete
- 80% of invoice volume auto-ingested
### Phase 2: Differentiators (Weeks 9-16)
| Week | Deliverable |
|------|-------------|
| 9-10 | Landed Cost Engine, dynamic pricing logic |
| 11-12 | Contract ingestion, discount validation, leakage reporting |
| 13-14 | TMS/WMS/ERP integration, planned vs. actual |
| 15-16 | Evidence generation, audit defense packages, dashboards |
**Exit Criteria:**
- Full landed cost calculation
- Contract-aware auditing
- Enterprise integrations operational
### Phase 3: Intelligence Layer (Weeks 17-24)
| Week | Deliverable |
|------|-------------|
| 17-18 | PRODUCT Module: Alcohol compliance matrix, excise calculation |
| 19-20 | FINANCE Module: Payment restrictions, tax nexus |
| 21-22 | Duty Recovery Service: Drawback automation, K32/E15 forms |
| 23-24 | ML anomaly detection, confidence model tuning |
**Exit Criteria:**
- 4 modules live
- Duty recovery operational
- 5+ customers across modules
### Phase 4: Platform Scale (Weeks 25-36)
| Week | Deliverable |
|------|-------------|
| 25-28 | Multi-tenant full deployment, self-service onboarding |
| 29-32 | US market expansion (USHTS, FDA, TTB rules) |
| 33-36 | Partner integrations (Shopify, WooCommerce apps), SOC 2 Type II |
**Exit Criteria:**
- Multi-country support
- Self-service customer onboarding
- SOC 2 certification
---
## 13. Financial Projections
### 13.1 Revenue Model
With proportional pricing (15% of savings, no caps):
| Year | Customers | Avg Modules | Avg Monthly | ARR |
|------|-----------|-------------|-------------|-----|
| Y1 | 50 | 1.5 | $8K | $4.8M |
| Y2 | 300 | 2.2 | $15K | $54M |
| Y3 | 1,000 | 2.8 | $22K | $264M |
| Y4 | 2,500 | 3.2 | $28K | $840M |
| Y5 | 5,000 | 3.5 | $30K | $1.8B |
### 13.2 Key Metrics Targets
| Metric | Y1 | Y3 | Y5 |
|--------|-----|-----|-----|
| Gross Margin | 75% | 82% | 85% |
| Net Revenue Retention | 130% | 145% | 150% |
| CAC Payback | 9 mo | 6 mo | 4 mo |
| LTV/CAC | 15x | 45x | 80x |
| Module Attach Rate | 1.5 | 2.8 | 3.5 |
### 13.3 Path to $1B Valuation
| Milestone | Timeline | ARR | Valuation |
|-----------|----------|-----|-----------|
| Seed Close | Q1 2026 | $0.5M | $5M |
| Series A | Q4 2026 | $4.8M | $60M |
| Series B | Q2 2028 | $54M | $500M |
| Series C | Q4 2029 | $180M | $1.5B |
| Exit Options | Q4 2030 | $500M | $5B |
---
## 14. Team & Execution
### 14.1 Current State
- **CEO/Founder:** Dan (14 years Zonos domain expertise, product vision, sales)
- **CTO:** Position open (David departed)
### 14.2 Critical Hire: Founding Engineer
**Profile:**
- Senior engineer from logistics/compliance space (Flexport, Shippo, Zonos, etc.)
- Full-stack capable, ships fast
- Comfortable with ambiguity
**Compensation:**
- 8-12% equity (CTO-level for non-CTO title)
- They own all technical decisions
- You focus on vision, sales, domain expertise
### 14.3 Execution Strategy: Manual First
While hiring:
- **Week 1:** Upgrade KnifeCenter to $2,788/month (deliver value manually)
- **Week 2:** Start Wayfair pilot at $30K/month potential
- **Month 1:** Hit $10K MRR before writing production code
- **Month 3:** Basic MVP while continuing manual delivery
**Principle:** Prove value with customers before engineering perfection.
---
## 15. Technical Architecture Detail
### 15.1 Monorepo Structure
```
rgl8r/
├── packages/
│ ├── @rgl8r/core # Universal compliance engine
│ ├── @rgl8r/ship # Shipping module
│ ├── @rgl8r/trade # Trade module
│ ├── @rgl8r/product # Product compliance module
│ ├── @rgl8r/finance # Finance module
│ ├── @rgl8r/ui # Design system
│ └── @rgl8r/api # API gateway
├── apps/
│ ├── web # rgl8r.com
│ ├── app # app.rgl8r.com
│ └── docs # docs.rgl8r.com
├── services/
│ ├── pcs # Product Compliance Service
│ ├── lce # Landed Cost Engine
│ ├── vas # Validation & Audit Service
│ ├── cds # Customs Documentation Service
│ ├── iel # Import/Export Ledger
│ ├── drs # Duty Recovery Service
│ ├── cre # Compliance Rules Engine
│ └── fas # Freight Audit Service
└── adapters/
├── knifecenter/ # KnifeCenter-specific integration
├── wayfair/ # Wayfair-specific integration
└── carriers/
├── fedex/
├── ups/
└── dhl/
```
### 15.2 Database Schema
```sql
-- Multi-tenant with row-level security
CREATE TABLE compliance_entities (
id UUID PRIMARY KEY,
tenant_id UUID NOT NULL,
module VARCHAR(50) NOT NULL,
entity_type VARCHAR(50) NOT NULL,
data JSONB NOT NULL,
created_at TIMESTAMP DEFAULT NOW(),
INDEX idx_tenant_module (tenant_id, module)
);
-- Validation results with evidence
CREATE TABLE compliance_results (
id UUID PRIMARY KEY,
entity_id UUID REFERENCES compliance_entities(id),
rule_id VARCHAR(100) NOT NULL,
status VARCHAR(20) NOT NULL,
confidence DECIMAL(3,2) NOT NULL,
evidence JSONB NOT NULL,
created_at TIMESTAMP DEFAULT NOW()
);
-- Attribute validation moat
CREATE TABLE attribute_validations (
id UUID PRIMARY KEY,
tenant_id UUID NOT NULL,
sku VARCHAR(100) NOT NULL,
attribute_name VARCHAR(100) NOT NULL,
original_value TEXT,
validated_value TEXT NOT NULL,
source VARCHAR(100) NOT NULL,
confidence DECIMAL(3,2) NOT NULL,
validated_at TIMESTAMP DEFAULT NOW(),
UNIQUE(tenant_id, sku, attribute_name)
);
-- CBSA tariff data
CREATE TABLE cbsa_tphs (
tariff VARCHAR(20) PRIMARY KEY,
desc1 TEXT,
desc2 TEXT,
mfn VARCHAR(20),
ceut VARCHAR(20),
ukt VARCHAR(20),
updated_at TIMESTAMP DEFAULT NOW()
);
-- Import/Export ledger (immutable)
CREATE TABLE customs_ledger (
id UUID PRIMARY KEY,
tenant_id UUID NOT NULL,
entry_type VARCHAR(20) NOT NULL, -- import, export, drawback
cad_number VARCHAR(50),
cers_number VARCHAR(50),
sku VARCHAR(100),
hs_code VARCHAR(20),
declared_value DECIMAL(12,2),
duty_paid DECIMAL(12,2),
tax_paid DECIMAL(12,2),
excise_paid DECIMAL(12,2),
refund_claimed DECIMAL(12,2),
variance_score VARCHAR(10),
documentation_hash VARCHAR(64),
created_at TIMESTAMP DEFAULT NOW()
);
-- Row-level security
CREATE POLICY tenant_isolation ON compliance_entities
USING (tenant_id = current_setting('app.current_tenant')::UUID);
```
### 15.3 API Architecture
```
api.rgl8r.com/v1/
├── /validate # Single-entity validation
├── /validate/batch # Bulk validation
├── /audit/freight # Freight invoice audit
├── /audit/customs # Customs classification audit
├── /classify # HS classification
├── /landed-cost # Landed cost calculation
├── /contracts # Contract management
├── /drawback # Duty recovery
├── /docs # Customs documentation
├── /ledger # Import/export records
├── /reports # Report generation
├── /webhooks # Event notifications
└── /admin # Tenant management
```
### 15.4 AI Agent & Automation Interfaces
RGL8R exposes multiple interfaces optimized for different automation contexts:
| Interface | Primary User | Discovery Method | Output Format |
|-----------|--------------|------------------|---------------|
| **REST API** | Traditional integrations, ERPs | OpenAPI spec, Swagger UI | JSON |
| **CLI** | DevOps, CI/CD pipelines | npm/brew, `--help` | JSON, exit codes |
| **MCP Server** | AI agents (Claude, etc.) | MCP registry, llms.txt | Structured tool results |
| **Webhooks** | Event-driven systems | API docs | JSON payloads |
**CLI Design Principles:**
```bash
# Semantic commands that map to business problems
rgl8r ship audit --invoices ./fedex-jan.csv --output json
rgl8r trade screen --catalog ./skus.csv --measures sima
rgl8r hs classify --description "stainless steel kitchen sink"
# Exit codes for pipeline integration
# 0 = success, no findings
# 1 = success, findings detected (action needed)
# 2 = error (bad input, auth failure, etc.)
```
**MCP Server Tools:**
```yaml
tools:
- name: screen_catalog_for_sima
description: |
Check if products being imported to Canada face anti-dumping
or countervailing duties under SIMA. Returns AT_RISK,
NEEDS_REVIEW, or CLEARED for each SKU.
parameters:
catalog_csv: CSV file with SKU, HS code, country of origin
returns:
findings: Array of {sku, outcome, exposure_amount, reason}
- name: audit_carrier_invoices
description: |
Analyze FedEx or UPS invoices for billing errors including
DIM weight variances, duplicate charges, and invalid accessorials.
parameters:
invoice_file: Carrier invoice export (CSV or XLSB)
carrier: fedex | ups
returns:
findings: Array of {tracking, type, variance_amount, confidence}
- name: estimate_landed_cost
description: |
Calculate total landed cost for importing a product to Canada,
including duty, GST/HST/PST, and any SIMA exposure.
parameters:
hs_code: 10-digit HS code
origin_country: ISO country code
value: Declared value in USD
destination_province: CA province code
returns:
landed_cost: {duty, tax, total, warnings}
```
**llms.txt Structure:**
```
# rgl8r.com/llms.txt
> RGL8R is a compliance operating system for e-commerce.
> It solves two problems:
> 1. Carriers overcharge on parcel invoices (DIM weight, duplicates, wrong accessorials)
> 2. Customs authorities penalize incorrect classifications (SIMA anti-dumping duties)
## What RGL8R Does
- Audits FedEx and UPS invoices to find billing errors (typically 3-8% of spend)
- Screens product catalogs for SIMA/anti-dumping exposure before importing to Canada
- Generates audit-defense evidence packages for customs inquiries
## Who Should Use RGL8R
- E-commerce brands shipping cross-border to Canada
- Retailers with $1M+ annual parcel spend
- Compliance teams managing HS classifications
## How to Access
- Web: app.rgl8r.com
- API: api.rgl8r.com (see OpenAPI spec)
- CLI: npm install -g @rgl8r/cli
- MCP: See mcp-registry for tool definitions
## API Documentation
- OpenAPI: api.rgl8r.com/docs
- Examples: docs.rgl8r.com/examples
```
---
## 16. Non-Functional Requirements
These are end-state platform requirements. 2026 will make pragmatic trade-offs, but architecture and logging should be designed with these targets in mind. We will not sell a formal 99.9% SLA in 2026; internally we will aim for high availability and good operational hygiene without contractual penalties.
| Category | Requirement |
|----------|-------------|
| Performance | <2s end-to-end landed-cost response |
| Scalability | Horizontally scalable (Kubernetes/ECS) |
| Availability | 99.9% uptime for pricing APIs |
| Security | SOC 2-aligned logging, PII masking, field-level encryption |
| Retention | 6 years + 1 month (CBSA requirement) |
| Auditability | Immutable event logs with payload hash (SHA-256) |
| Localization | Multi-currency, EN/FR for Canada |
| Extensibility | Plug-in rulesets, new markets without code changes |
### 16.1 Compliance Posture
**Important for customers, investors, and legal review:**
RGL8R provides validation, decision support, workflow automation, and audit-ready evidence. **We do not act as a customs broker or submit filings on behalf of customers.**
| What RGL8R Does | What RGL8R Does NOT Do |
|-----------------|------------------------|
| Validates HS classifications with confidence scores | Make final classification decisions for customers |
| Screens for SIMA/ADD exposure and flags risk | Guarantee duty rates or customs outcomes |
| Generates evidence blocks for audit defense | Submit customs declarations or B3 forms |
| Provides landed cost estimates and variance detection | Act as importer of record |
| Stores audit trail with timestamps and provenance | Replace licensed customs brokers |
**Customer responsibility:** Importer-of-record obligations remain with the customer. Customers are responsible for final declarations, broker selection, and regulatory submissions.
**Our commitment:** We log all decisions and evidence for defensibility. When customers face CBSA inquiries, they can export audit defense packages with full provenance chains.
---
## 17. Success Metrics
### 17.1 North Star
**Compliance Value Generated** — Total $ saved/prevented across all customers
### 17.2 Module Metrics
| Module | Primary Metric | Target |
|--------|----------------|--------|
| SHIP | Freight spend recovered | 3-5% |
| TRADE | Classification accuracy | ≥95% |
| TRADE | Landed cost variance | ≤1% |
| SIMA | Exposure prevented | $0 unexpected |
| PRODUCT | Compliance violations | 0 |
| Overall | Customer ROI | ≥5x |
### 17.3 Platform Metrics
- Net revenue retention: >140%
- Module attach rate: 2.5+ modules/customer
- Time to first value: <24 hours
- Classification auto-rate: >90%
- Audit pass rate: 100%
---
## 18. Differentiation Summary (Your Moat)
RGL8R wins because it:
1. **Unifies compliance domains** — One platform for Ship, Trade, Finance, Product, Geographic
2. **Has attribute validation IP** — Every validation improves the next; competitors start from zero
3. **Provides audit defense** — Evidence blocks and regulatory references enable customs audit survival
4. **Handles messy reality** — Built for real catalogs with missing attributes, not pristine datasets
5. **Calculates true landed cost** — Internal engine + external validation = ≤1% variance
6. **Automates duty recovery** — Drawback, destruction refunds, K32/E15 automation
7. **Scales globally** — Country = config folder, not code rewrite
8. **Aligns incentives perfectly** — 15% of savings means everyone wins at 6.7x ROI
**This is not an "audit tool" or "classification service."**
**This is regulatory intelligence infrastructure for commerce.**
---
## 19. What to Generate Next
From this blueprint, I can create:
- **PRDs** — Detailed product requirements by epic
- **UI Wireframes** — Screen-by-screen layouts
- **Database Schema** — Full Prisma/SQL definitions
- **API Specifications** — OpenAPI/Swagger docs
- **Rule Engine Design** — Configuration format, execution model
- **EDI Implementation Guide** — X12 parsing, mapping, acknowledgments
- **Integration Playbooks** — ERP/TMS/WMS specific guides
- **Landed Cost Engine Spec** — Calculation logic, tax tables, trade agreements
- **Duty Recovery Playbook** — Drawback process, K32/E15 automation
- **Go-to-Market Materials** — Positioning, competitive analysis
- **Investor Narrative** — Pitch deck content, financial model inputs
---
**This blueprint is your foundation.**
**RGL8R: We enable compliant commerce.**
---
**Document Version:** 3.8
**Last Updated:** February 2026
**Status:** Strategy Document — See `specs/` for frozen implementation details
**Aligned With:** Dashboard UX Spec v2.6.9, Schema v1.9, Self-Extending Architecture v1
**v3.7 Changes:**
- Added §8.3: AI Agent Discovery Channel (GTM strategy for AI-assisted discovery)
- Added §10.5: AI Agents & Automation Partners (user stories for CLI, MCP, agent discovery)
- Added §15.4: AI Agent & Automation Interfaces (CLI design, MCP tools, llms.txt structure)
**v3.8 Changes:**
- Added corridor-agnostic architecture: corridor as first-class parameter throughout
- Added TRADE Module: Export Controls (ITAR/EAR) and Denied Party Screening (Phase 2)
- Updated §9.1: Architecture Readiness with corridor-aware rule inheritance model
- Updated §9.3: Multi-Country Rule Structure with full scope hierarchy (universal → origin → destination → corridor → party → product → finance → carrier)
- Updated §11: Pain Point → Solution Mapping with export controls and denied party screening
- Explicit Phase 1 scope: architecture is corridor-agnostic, rules are Canada-focused
- Cross-reference to `strategy/self-extending-architecture-v1.md` for rule inheritance details