net/docs/CONTENT_FACTORY_PLAN.md

🏭 Spark Content Factory - Implementation Plan

Overview

Transform the three intelligence files into a fully automated content generation system that creates hyper-personalized articles by combining:

• WHO (Avatar + Niche)
• WHERE (City + Wealth Cluster)
• WHAT (Offer Block + Spintax)

---

📊 Architecture Diagram

┌─────────────────────────────────────────────────────────────────────┐
│                         DIRECTUS SCHEMA                              │
├─────────────────────────────────────────────────────────────────────┤
│                                                                      │
│  ┌─────────────┐     ┌─────────────┐     ┌─────────────┐            │
│  │   SITES     │────▶│  CAMPAIGNS  │────▶│  ARTICLES   │            │
│  │ (Your Sites)│     │(What to build│     │(Generated)  │            │
│  └─────────────┘     └─────────────┘     └─────────────┘            │
│         │                   │                    │                   │
│         │            ┌──────┴──────┐             │                   │
│         │            ▼             ▼             ▼                   │
│         │     ┌───────────┐ ┌───────────┐ ┌───────────┐             │
│         │     │  AVATARS  │ │  NICHES   │ │ LOCATIONS │             │
│         │     │ (Who)     │ │ (Industry)│ │ (Where)   │             │
│         │     └───────────┘ └───────────┘ └───────────┘             │
│         │            │             │             │                   │
│         │            └──────┬──────┘             │                   │
│         │                   ▼                    │                   │
│         │           ┌─────────────┐              │                   │
│         └──────────▶│OFFER BLOCKS │◀─────────────┘                   │
│                     │(Messaging)  │                                  │
│                     └─────────────┘                                  │
│                            │                                         │
│                            ▼                                         │
│                     ┌─────────────┐                                  │
│                     │ SEO ENGINE  │                                  │
│                     │• Meta Title │                                  │
│                     │• Meta Desc  │                                  │
│                     │• Schema.org │                                  │
│                     └─────────────┘                                  │
│                                                                      │
└─────────────────────────────────────────────────────────────────────┘

---

📁 Directus Collections to Create

1. avatars (FROM: avatar_intelligence.json)

Field	Type	Description
id	uuid	Primary key
slug	string	scaling_founder, elite_consultant, etc.
base_name	string	"The Tech Titan / Scaling Founder"
wealth_cluster	string	"Tech-Native"
psychographics	text	Long description of mindset
tech_stack	json	["Zapier", "Slack", "AWS"]
pronoun_male	string	"he"
pronoun_female	string	"she"
identity_male	string	"bottlenecked technical founder"
identity_female	string	"bottlenecked technical founder"

2. niches (FROM: avatar_intelligence.json → business_niches)

Field	Type	Description
id	uuid	Primary key
name	string	"Vertical SaaS (B2B)"
slug	string	"vertical-saas-b2b"
avatar	m2o → avatars	Which avatar owns this niche
keywords	json	SEO keywords for this niche
pain_points	json	Common pains in this niche

3. wealth_clusters (FROM: geo_intelligence.json)

Field	Type	Description
id	uuid	Primary key
slug	string	tech_native, financial_power
name	string	"The Silicon Valleys"
tech_adoption_score	integer	1-10
primary_need	string	"Advanced Custom Automation & SaaS"
matching_avatars	m2m → avatars	Which avatars match this cluster

4. elite_cities (FROM: geo_intelligence.json → cities)

Field	Type	Description
id	uuid	Primary key
name	string	"Atherton"
state	string	"CA"
full_name	string	"Atherton, CA"
wealth_cluster	m2o → wealth_clusters	Which cluster
landmarks	json	Local landmarks for spintax

5. offer_blocks (FROM: offer_engine.json)

Field	Type	Description
id	uuid	Primary key
slug	string	block_01_zapier_fix
title	string	"The $1,000 Fix"
hook	text	"Stop the bleeding in your {{NICHE}} business."
spintax	text	Full spintax template
avatar_pains	json	{ avatar_slug: [pain1, pain2, pain3] }
meta_title_template	string	"{{OFFER}} for {{NICHE}} in {{CITY}}"
meta_desc_template	text	SEO description template

6. content_campaigns (User creates these)

Field	Type	Description
id	uuid	Primary key
site	m2o → sites	Which site to publish to
name	string	"Q1 2025 - Tech Founders"
target_avatars	m2m → avatars	Which avatars to target
target_niches	m2m → niches	Which niches
target_cities	m2m → elite_cities	Which cities
offer_blocks	m2m → offer_blocks	Which offers to use
velocity_mode	select	RAMP_UP, STEADY, SPIKES
target_count	integer	How many articles

7. generated_articles (Factory output)

Field	Type	Description
id	uuid	Primary key
site	m2o → sites	Published to this site
campaign	m2o → content_campaigns	Source campaign
avatar	m2o → avatars	Target avatar
niche	m2o → niches	Target niche
city	m2o → elite_cities	Target city
offer	m2o → offer_blocks	Offer used
headline	string	Generated headline
meta_title	string	SEO title (60 chars)
meta_description	string	SEO desc (160 chars)
full_html_body	text	The article content
schema_json	json	Schema.org markup
sitemap_status	select	ghost, queued, indexed
date_published	datetime	Backdate or now

---

🔄 How It All Connects

Page Generation Flow

USER SELECTS:
┌─────────────────────────────────────────────┐
│ Site: la.christopheramaya.work              │
│ Avatar: scaling_founder                     │
│ Niche: Vertical SaaS (B2B)                  │
│ City: Palo Alto, CA                         │
│ Offer: The $1,000 Fix                       │
│ Count: 50 articles                          │
└─────────────────────────────────────────────┘
                    │
                    ▼
FACTORY GENERATES:
┌─────────────────────────────────────────────┐
│ FOR EACH COMBINATION:                       │
│                                             │
│ 1. Pull avatar psychographics               │
│ 2. Pull niche-specific pains                │
│ 3. Pull city landmarks                      │
│ 4. Pull offer spintax                       │
│ 5. Replace all {{TOKENS}}                   │
│ 6. Spin the spintax                         │
│ 7. Generate SEO meta                        │
│ 8. Create schema.org JSON                   │
│ 9. Save to generated_articles               │
│ 10. Apply Gaussian scheduling               │
└─────────────────────────────────────────────┘

Token Replacement Map

Token	Source	Example
{{NICHE}}	niches.name	"Vertical SaaS"
{{CITY}}	elite_cities.name	"Palo Alto"
{{STATE}}	elite_cities.state	"CA"
{{AVATAR}}	avatars.identity_male	"bottlenecked technical founder"
{{PRONOUN}}	avatars.pronoun_male	"he"
{{TECH_STACK}}	avatars.tech_stack[random]	"Zapier"
{{LANDMARK}}	elite_cities.landmarks[random]	"Stanford University"
{{AGENCY_NAME}}	sites.name	"Spark Digital"
{{AGENCY_URL}}	sites.domain	"sparkdigital.com"
{{CURRENT_YEAR}}	context	"2024"
{{WEALTH_VIBE}}	wealth_clusters.primary_need	"Advanced Custom Automation"

---

📋 SEO Meta Generation

For each article, auto-generate:

Meta Title (60 chars)

{{OFFER_TITLE}} for {{NICHE}} Businesses in {{CITY}}, {{STATE}}

Example: "The $1,000 Fix for Vertical SaaS Businesses in Palo Alto, CA"

Meta Description (160 chars)

{{AVATAR_IDENTITY}} in {{CITY}}? {{OFFER_HOOK}} We {{SOLUTION}}. Get your free audit today.

Example: "Bottlenecked technical founder in Palo Alto? Stop the bleeding in your SaaS business. We rebuild broken automation. Get your free audit today."

Schema.org JSON-LD

{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "{{META_TITLE}}",
  "description": "{{META_DESC}}",
  "author": {
    "@type": "Organization",
    "name": "{{AGENCY_NAME}}"
  },
  "datePublished": "{{DATE_PUBLISHED}}",
  "dateModified": "{{DATE_MODIFIED}}",
  "publisher": {
    "@type": "Organization",
    "name": "{{AGENCY_NAME}}"
  }
}

---

🚀 User Workflow in Directus

Step 1: Add Your Site

Sites → + New
- Name: "Spark Digital LA"
- Domain: "la.christopheramaya.work"

Step 2: Create Campaign

Content Campaigns → + New
- Site: (dropdown) Spark Digital LA
- Target Avatars: ☑️ scaling_founder ☑️ saas_overloader
- Target Niches: ☑️ Vertical SaaS ☑️ Fintech
- Target Cities: ☑️ Palo Alto ☑️ Austin ☑️ Seattle
- Offer Blocks: ☑️ Zapier Fix ☑️ Market Domination
- Velocity: RAMP_UP
- Target Count: 100

Step 3: Click "Generate"

→ Factory creates 100 unique articles
→ Each article = unique combo
→ SEO meta auto-generated
→ Gaussian scheduling applied

Step 4: Review & Publish

Generated Articles → Filter by Campaign
→ Preview any article
→ Approve test batch
→ Click "Publish to Site"
→ Articles go live

---

📊 Combination Math

With full data:

• 10 Avatars × 10 Niches each = 100 Avatar-Niche combos
• 50 Elite Cities
• 10 Offer Blocks

Maximum unique articles: 100 × 50 × 10 = 50,000 pages

For a focused campaign:

• 2 Avatars × 3 Niches × 10 Cities × 2 Offers = 120 articles

---

✅ Implementation Tasks

Phase 1: Schema Setup

• Create avatars collection
• Create niches collection
• Create wealth_clusters collection
• Create elite_cities collection
• Create offer_blocks collection
• Update content_campaigns with relations
• Update generated_articles with relations

Phase 2: Data Import

• Import 10 avatars
• Import 100 niches (10 per avatar)
• Import 5 wealth clusters
• Import 50 elite cities
• Import offer blocks

Phase 3: Factory Engine

• Update token processor
• Build campaign generator
• Add SEO meta templates
• Add schema.org generator

Phase 4: Testing

• Generate test batch
• Verify token replacement
• Verify SEO meta quality