Skip to main content

​
Publisher Business Unit

Status: Documentation
Owner: Jeff Liu
Linear Project: PRD (Publisher)

​
What Publisher Does

Publisher is the multi-vertical blog network - 50+ industries producing SEO-optimized content at scale with automated affiliate revenue. 
Content Sources → Intelligence Layer → Generated Articles → Distribution → Revenue
     ↓                    ↓                    ↓                ↓           ↓
  Competitors         Patterns            SEO-optimized      50+ sites   Affiliates
  Social feeds        Scoring             Trending topics    Syndication  Ads
  News/RSS           Generation           Multi-format       Email       Sponsors

​
How Publisher Uses the Intelligence Layer

Publisher is the FIRST consumer of the shared intelligence layer. Everything built here gets extracted to packages/ for Platform reuse.
Intelligence FeaturePublisher Use Case
Content AnalysisExtract topics, keywords, facts from scraped content
Pattern Matching”This topic format works for tech vertical”
Content ScoringPrioritize which content to produce
GenerationSEO articles, social posts, email
DistributionPublish to verticals, syndicate
Feedback LoopWhich articles drive traffic/revenue

​
Publisher Tables (12)

TablePurposeKey Fields
verticalsIndustry categoriesslug, domain, seo_config
sitesIndividual blogsvertical_id, domain, theme
articlesPublished contentsite_id, seo_metadata, affiliate_links
seo_trackingSERP monitoringarticle_id, keyword, position
affiliate_linksTrackable linksprovider, product_id, commission
affiliate_clicksClick trackinglink_id, visitor_id, converted
ad_placementsAd inventorysite_id, position, provider
ad_revenueRevenue trackingplacement_id, impressions, revenue
authorsContent attributionname, bio, avatar_url
editorial_calendarContent planningvertical_id, publish_date, status
content_briefsArticle specskeywords, outline, target_length
syndicationCross-postingarticle_id, target_platform, status

​
Key Flows

​
1. Content Production Pipeline

1. Scrape competitor content (content_sources)
2. Analyze for topics/keywords (content_analysis)
3. Score against vertical patterns (content_platform_scores)
4. Generate brief for high-scoring topics (content_briefs)
5. Generate article (generated_content)
6. Publish to site (articles)
7. Track SEO performance (seo_tracking)
8. Measure revenue (affiliate_clicks, ad_revenue)

​
2. Vertical Selection

23 verticals selected by:
  • Search volume (monthly searches)
  • CPC (advertiser willingness to pay)
  • Affiliate potential (commission rates)
  • Content velocity (how fast topics change)
  • Competition (domain authority needed)

​
3. Affiliate Revenue Flow

Article with affiliate links
       ↓
Reader clicks → affiliate_clicks recorded
       ↓
Purchase on merchant site
       ↓
Commission webhook → affiliate_clicks.converted = true
       ↓
Revenue attributed to article/vertical

​
Subdomain Architecture

Each vertical gets a subdomain:
  • tech.trendingsociety.com
  • travel.trendingsociety.com
  • realestate.trendingsociety.com
Benefits:
  • Unified SEO authority (trendingsociety.com domain)
  • Vertical-specific theming
  • Shared infrastructure
  • Consolidated analytics
DocumentWhat It Covers
SCHEMA.mdPublisher table definitions
SCHEMA_CONCEPTS.mdIntelligence layer (used by Publisher)
verticals.mdVertical selection criteria (coming)
seo-engine.mdSERP tracking system (coming)
affiliate-system.mdRevenue attribution (coming)
editorial-flow.mdBrief → Article pipeline (coming)
syndication.mdCross-posting strategy (coming)

​
Linear Issues

Publisher work uses the PRD project prefix in Linear. Example issue structure: 
## Context
Link to: docs/publisher/OVERVIEW.md
Link to: SCHEMA_CONCEPTS.md#content-intelligence-loop

## What
[Specific deliverable]

## Acceptance Criteria
- [ ] Criteria 1
- [ ] Criteria 2

## Cursor Prompt (if delegate:cursor)
[Specific implementation instructions]
See: Linear Issue Standards for full templates.