# INVESTOR BRIEF: Spark Platform > **BLUF**: Spark is a multi-tenant content scaling platform that generates location-targeted SEO articles using spintax permutation and Cartesian pattern matching. Current capacity: 50,000+ unique articles per campaign configuration. --- ## 1. Platform Function Spark automates SEO content production at scale. The system: 1. Ingests buyer personas (Avatars), location data, and content templates 2. Computes Cartesian products of location × persona × offer variations 3. Generates unique articles with geo-specific and persona-specific content 4. Manages multi-site content distribution with scheduling controls **Core Problem Solved**: Manual SEO content creation produces ~2-5 articles/day. Spark produces 100-500 articles/hour with equivalent uniqueness and targeting precision. --- ## 2. Technical Stack | Layer | Technology | Version | Purpose | |-------|------------|---------|---------| | Frontend | Astro | 4.7 | SSR + Islands Architecture | | UI | React | 18.3 | Interactive components | | Backend | Directus | 11 | Headless CMS + REST/GraphQL | | Database | PostgreSQL | 16 | Primary data store | | Extensions | PostGIS | 3.4 | Geographic queries | | Cache | Redis | 7 | Session + job queue backing | | Queue | BullMQ | - | Async job processing | | Deployment | Coolify | - | Docker orchestration | **Infrastructure**: Self-hostable via Docker Compose. No external SaaS dependencies. --- ## 3. Data Assets ### 3.1 Location Database | Dataset | Count | Source | |---------|-------|--------| | US States | 51 | Census data | | US Counties | 3,143 | Census data | | US Cities | ~50 per county | Population-ranked | ### 3.2 Intelligence Assets | Asset Type | Description | IP Value | |------------|-------------|----------| | Avatar Intelligence | 10+ buyer personas with psychographics | Proprietary | | Wealth Clusters | 5+ economic profile groupings | Proprietary | | Spintax Dictionaries | Word/phrase variation libraries | Proprietary | | Cartesian Patterns | Title/hook formula combinations | Proprietary | | Offer Blocks | 10+ promotional templates | Proprietary | --- ## 4. Capacity Metrics ### 4.1 Theoretical Maximum ``` 10 Avatars × 10 Niches × 50 Cities × 10 Offers = 50,000 unique articles ``` ### 4.2 Practical Campaign ``` 2 Avatars × 3 Niches × 10 Cities × 2 Offers = 120 articles ``` ### 4.3 Generation Speed | Operation | Throughput | |-----------|------------| | Headline permutation | 1,000/second | | Article assembly | 100-500/hour | | Queue processing | Configurable batch size | --- ## 5. Architecture Overview ``` ┌─────────────────────────────────────────────────────────────────┐ │ USER REQUEST │ └──────────────────────────┬──────────────────────────────────────┘ ▼ ┌─────────────────────────────────────────────────────────────────┐ │ TRAEFIK (Reverse Proxy) │ │ Routes by domain/path to containers │ └──────────────────────────┬──────────────────────────────────────┘ ▼ ┌──────────────────┴──────────────────┐ ▼ ▼ ┌───────────────┐ ┌───────────────┐ │ FRONTEND │ │ DIRECTUS │ │ Astro SSR │◄──── REST API ────►│ Port 8055 │ │ Port 4321 │ │ Headless CMS │ └───────────────┘ └───────┬───────┘ │ ▼ ┌───────────────────────────────────┐ │ POSTGRESQL 16 + POSTGIS │ │ 30+ Collections │ │ Harris Matrix Schema Order │ └───────────────────┬───────────────┘ │ ▼ ┌───────────────────────────────────┐ │ REDIS 7 │ │ Session Cache + BullMQ Jobs │ └───────────────────────────────────┘ ``` --- ## 6. Multi-Tenancy Model | Isolation Level | Implementation | |-----------------|----------------| | Data | `site_id` FK on all content tables | | Routes | Domain-based routing via Traefik | | Authentication | Directus role-based access control | | Permissions | Site-scoped API tokens | **Tenant capacity**: Unlimited sites. Horizontal scaling via Docker replicas. --- ## 7. Feature Inventory ### 7.1 Intelligence Library (Data Management) - Avatar Intelligence Manager - Geo Intelligence Map (Leaflet integration) - Spintax Dictionary Manager - Cartesian Pattern Builder ### 7.2 Content Factory (Production) - Kanban workflow board - Campaign wizard (geo + spintax modes) - Jobs queue with progress monitoring - Scheduler with Gaussian distribution ### 7.3 Launchpad (Site Builder) - Multi-site management - Block-based page builder - Navigation editor - Theme customization ### 7.4 SEO Engine (Optimization) - Headline generation (spintax permutation) - Fragment assembly (6-pillar structure) - Internal link insertion - Duplicate content detection - Sitemap drip scheduling ### 7.5 Analytics (Tracking) - Pageview tracking - Event tracking - Conversion tracking - Dashboard with aggregations --- ## 8. Competitive Position | Capability | Spark | Generic CMS | AI Writers | |------------|-------|-------------|------------| | Multi-tenant | ✓ | Partial | ✗ | | Geo-targeting | ✓ (3,143 counties) | ✗ | ✗ | | Persona-targeting | ✓ (10+ avatars) | ✗ | Limited | | Spintax processing | ✓ | ✗ | ✗ | | Cartesian patterns | ✓ | ✗ | ✗ | | Self-hostable | ✓ | Varies | ✗ | | Queue-based scaling | ✓ | ✗ | ✗ | --- ## 9. Revenue Model Indicators | Model | Mechanism | |-------|-----------| | SaaS per site | Monthly fee per managed site | | Volume pricing | Tiered by articles generated | | Enterprise | Self-hosted license + support | | White-label | Reseller partnerships | --- ## 10. Development Status | Milestone | Status | Description | |-----------|--------|-------------| | M1: Intelligence Library | ✓ Complete | Full CRUD + stats | | M2: Content Factory | ✓ Complete | Kanban + queue | | M3: All Collections | ✓ Complete | 30+ schemas | | M4: Launchpad | ✓ Complete | Site builder | | M5: Production | ✓ Deployed | Live on Coolify | **Current Status**: Operational. Active development on feature enhancements. --- ## 11. Key Files | File | Purpose | |------|---------| | `complete_schema.sql` | Golden Schema (Harris Matrix ordered) | | `docker-compose.yaml` | Infrastructure definition | | `frontend/src/lib/schemas.ts` | TypeScript type definitions | | `frontend/src/pages/api/*` | 30+ API endpoints |