Claude-skill-registry campaign-structure
Design optimal Facebook/Meta campaign structure for testing including ABO vs CBO, ad set organization, targeting, and budget allocation. Use when setting up new campaigns, planning test structure, or optimizing account architecture.
install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/campaign-structure" ~/.claude/skills/majiayu000-claude-skill-registry-campaign-structure && rm -rf "$T"
manifest:
skills/data/campaign-structure/SKILL.mdsource content
Campaign Structure Designer
Create optimal campaign architecture for testing and scaling.
Process
Step 1: Assess Offer and Budget
Offer Type Assessment
- VSL/long-form funnel → More controlled testing
- Lead gen/simple funnel → Can be more aggressive
- New offer → Need more testing budget
- Proven offer → Can scale faster
Budget Allocation
- Daily testing budget
- Target CPA/CPL
- Acceptable loss threshold
- Timeline to profitability
Step 2: Choose Structure Type
ABO Testing Structure (Recommended for Testing) Use when: Finding initial winners, testing new angles
Campaign: [Offer] Testing ├── Ad Set 1: [Concept A] - $X/day │ ├── Ad 1: Hook variation 1 │ ├── Ad 2: Hook variation 2 │ └── Ad 3: Hook variation 3 ├── Ad Set 2: [Concept B] - $X/day │ ├── Ad 1... ...
CBO Scaling Structure Use when: Scaling proven winners
Campaign: [Offer] Scale - $X/day CBO ├── Ad Set 1: Winners Only │ ├── Winner Ad 1 │ ├── Winner Ad 2 │ └── Winner Ad 3
2x2x3 Testing Method (Jason K)
- 2 Headlines
- 2 Ad texts
- 3 Images/faces = 12 ad variations testing copy combinations
Double Six Shooter (Jason K)
- 10-12 ad sets (different avatars)
- Same 10-12 images per ad set
- Large CBO with minimum spend per ad set
- Let Facebook find winning avatar + image combos
Step 3: Set Targeting Parameters
Testing Phase
- 2% Lookalike audiences (Jason K recommends)
- Mobile newsfeed only
- WiFi only (for VSL)
Scaling Phase
- 10% Lookalike (Jason K)
- All placements
- Broader audiences
Audience Sources
- Purchase lookalikes
- Add to cart lookalikes
- Email list lookalikes
- Page engagers (for exclusions)
Step 4: Define Budget Rules
Testing Budget
- Spend 1.5-2x target CPL before calling winner (Casto)
- Use initiate checkout as leading indicator (3x purchase data)
- Kill at threshold, don't hold hoping
Winner Criteria (Jason K)
- Doesn't lose more than once in 3 days
- Doesn't lose more than twice in 7 days
- Consistent performance, not just spikes
Scaling Budget
- 20% increases (safe)
- Can jump when hot, but risk breaking
- Spread across multiple ad accounts
Step 5: Output Campaign Blueprint
## CAMPAIGN STRUCTURE: [Offer Name] ### PHASE 1: INITIAL TESTING **Campaign: [Offer] - ABO Test** - Objective: Conversions - Budget: $[X]/day per ad set - Duration: [X] days minimum **Ad Set Structure:** | Ad Set | Concept | Targeting | Budget | |--------|---------|-----------|--------| | 1 | [Concept A] | 2% LAL | $X | | 2 | [Concept B] | 2% LAL | $X | | 3 | [Concept C] | 2% LAL | $X | **Per Ad Set:** - 5-10 ads per concept - Mix of hooks/variations - Mobile newsfeed only **Kill Criteria:** - CPL > [X] after $[spend] - CPA > [X] after [conversions] **Winner Criteria:** - CPL < [target] - Consistent 3+ days - Scalable volume --- ### PHASE 2: WINNER VALIDATION **Campaign: [Offer] - CBO Scale** - Budget: $[X]/day CBO - Winners only (3-5 ads) - Scale in place initially **Expansion:** - Replicate to Ad Account 2 - Replicate to Ad Account 3 - De-risk across accounts --- ### PHASE 3: FULL SCALE **Structure:** - Multiple CBOs across accounts - $10K max per CBO (avoid instability) - Weekend campaigns (separate) - Day-parting: 4AM-1PM Eastern (VSL) **Targeting Expansion:** - Move to 10% LAL - Test broad - All placements --- ### CREATIVE TESTING CADENCE **Weekly:** - 50-100 new creative variations - New hooks on proven angles - New angles on proven offers **Monthly:** - New concept batches - Founder content refresh - Seasonal angles ### TRACKING SETUP - UTM structure: [Define] - Conversion events: [List] - Attribution window: [Setting]
Key Principles
From Casto:
- Broad targeting, let creative do targeting
- Use bid caps/cost caps when margins thin
- Spread across multiple ad accounts
- Build internal tools for launch automation
From Jason K:
- 2% LAL for testing, 10% for scaling
- Watch initiate checkouts (3x data)
- Day-part for VSL (turn off afternoon)
- Weekend campaigns with higher budgets
- Large CBO of winners is end goal
Source: Jason K, Casto (Meta-CastovsJasonK)