Systems Lab

Agent skill

meta-campaign-structure-advisor

This skill should be used when the user asks to \"set up campaign structure\", \"audit account structure\", \"organize ad sets\", or mentions \"Meta account structure\", \"creative testing framework\", or \"4-tier funnel\".

activeSelf-containedActs undeclared2,661 words

Filed under Pricing and packaging.

From Ad-Superpowers/ad-superpowers-plugin · 120 skills · 5 · pushed 2026-09-10

What it does when it runs

This skill should be used when the user asks to \"set up campaign structure\", \"audit account structure\", \"organize ad sets\", or mentions \"Meta account structure\", \"creative testing framework\", or \"4-tier funnel\". Do NOT use for: bid strategy selection (use bid-strategy-selector), creative brainstorming (use creative-diversification-generator), full-funnel design across platforms (use full-funnel-designer).

Read from the skill and the 2 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply
None found.
Hosts it reaches
No third-party host appears in the skill or its bundled files.
Tool permissions it declares
No allowed-tools in the frontmatter. It does act, so it runs under whatever permissions your session already grants.
Actions present in the files
writes files

Ask about meta-campaign-structure-advisor

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?ClaudeChatGPT
Adapt it to my stackClaudeChatGPT
What else do I need for it to workClaudeChatGPT
Rather ask a human? Talk to Cheetah
git clone --depth 1 --filter=blob:none --sparse https://github.com/Ad-Superpowers/ad-superpowers-plugin.git /tmp/ad-superpowers-plugin
git -C /tmp/ad-superpowers-plugin sparse-checkout set "plugin/skills/meta-campaign-structure-advisor"
mkdir -p ~/.claude/skills/meta-campaign-structure-advisor
cp -R "/tmp/ad-superpowers-plugin/plugin/skills/meta-campaign-structure-advisor/." ~/.claude/skills/meta-campaign-structure-advisor/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

Or take the whole library

This repo ships a .claude-plugin manifest, so Claude Code can install all 120 skills at once. Plugin skills are invoked as /<plugin>:<skill>, so they never collide with your own.

/plugin marketplace add Ad-Superpowers/ad-superpowers-plugin
/plugin

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Reproduced in full from Ad-Superpowers/ad-superpowers-plugin/blob/9b6385d2d2d228e4dac096a1d6bc5715c04fa736/plugin/skills/meta-campaign-structure-advisor/SKILL.md, which is licensed MIT (repository). 2,661 words, 55 headings.

Meta Ads Campaign Structure Advisor

Framework for setting up and optimizing Meta Ads account structures using the 4-tier funnel methodology. Based on Meta's Performance 5 framework and proven agency best practices 2025-2026.

April 2026 API Update: Meta API v25.0 (Feb 2026) unifies Advantage+ Sales Campaigns (ASC) into a single campaign type with 3 automation levers (budget, audience, placement). The legacy smart_promotion_type / AAC distinction is deprecated (deadline May 19, 2026). advantage_state_info is the new field. In practice, the 4-tier structure below remains valid — ASC is the SCALE campaign type. New placement: Threads (400M+ MAU, March 2026) — add to placement mix for scale campaigns.

Core Principle

Creative = Targeting. Meta's Andromeda algorithm selects ads based on creative quality, not audience parameters. Creative diversification and systematic testing are the primary growth levers.

Simplification = Better Performance. Accounts with 2-4 campaigns perform 32% better on CPA than fragmented structures.

How this maps to our MCP tools today

Before applying the structures below, know what meta_create actually does today:

  • Campaign creation always requires a budget, so every campaign created via the tool is CBO (the budget lives on the campaign). Per-ad-set budgets (ABO) are not creatable through the tool yet — model ABO splits in Ads Manager if you need them.
  • Under CBO, ad sets must not carry their own budget — omit it; they share the campaign budget.
  • The funnel below is the strategy; for the click-by-click create sequence (campaign → ad set → ads), use ad-launch-playbook.

4-Tier Funnel Structure

TierCampaign TypeBudgetGoal
SCALEAdvantage+ Sales (ASC)60-70%Scale proven winners
TESTManual Sales (ABO)10-20%Identify new winners
RETARGETManual Sales15-25%Convert warm traffic
RETAINManual Sales5-10%Maximize customer LTV

SCALE Campaign (ASC) - 60-70% budget

  • Targeting: 100% Broad (Advantage+ Audience)
  • Creatives: Only validated winners - no new tests here
  • Goal: Maximum scale with proven performers

TEST Campaign (Prospecting) - 10-20% budget

  • Broad Pack [date] - Contains 3-5 new creatives for testing
  • Broad Pack [date] - New batch of creatives (weekly/biweekly)
  • Interest #1 Winners - Receives graduated winners for interest targeting
  • Interest #2 Winners - Receives graduated winners for other interests
  • Goal: Systematically test new creatives and identify winners

RETARGET Campaign - 15-25% budget

  • 14 Day FB/IG Engagers - Social engagement retargeting
  • 30 Day Website Visitors - Site visitor retargeting
  • 90 Day ATC/IC - Add to Cart / Initiate Checkout retargeting
  • Goal: Convert warm traffic with high purchase intent

RETAIN Campaign - 5-10% budget

  • All Time Purchasers - Upsell/Cross-sell messaging
  • 180 Day Purchasers - Replenishment/New Products messaging
  • Goal: Reactivate existing customers and increase LTV

Creative Flow: How Ads Move Through the System

This is the exact flow of how creatives are tested, evaluated, and graduated.

Step 1: New Creatives Start in TEST

All new creatives begin in the TEST campaign, specifically in a Broad Pack ad set:

New creative made
         ↓
    TEST Campaign
         ↓
  Broad Pack [date]
    (Ad Set with ABO)

Why Broad Pack?

  • Broad targeting gives the algorithm maximum freedom to find the best audience
  • ABO (Ad Set Budget Optimization) ensures equal budget distribution between creatives
  • Date in name makes tracking creative batches simple

Step 2: Evaluation Against Winner Criteria

After sufficient spend and time, each creative is evaluated:

Creative in Broad Pack
         ↓
  Spend >= minimum threshold?
  (e.g., 3x CPA or 3x AOV)
         ↓
    NO → Keep testing OR kill if underperforming
    YES → Evaluate performance
         ↓
  ROAS >= threshold AND conversions >= minimum?
         ↓
    NO → Kill or keep testing
    YES → WINNER! Go to Step 3

Step 3: Winner Graduation (Three Destinations)

Important: A winner goes to MULTIPLE destinations simultaneously, not just one.

When a creative reaches winner status, it is copied to:

Destination A: SCALE Campaign (ASC)

Winner Creative
      ↓
 SCALE Campaign (ASC)
      ↓
 100% Broad Ad Set
 (alongside other winners)
  • Creative is added to the ASC campaign
  • Runs with 100% broad targeting on Advantage Campaign Budget
  • Receives the largest share of budget (60-70%)

Destination B: Interest #1 Winners Ad Set

Winner Creative
      ↓
  TEST Campaign
      ↓
Interest #1 Winners
   (Ad Set)
  • Same creative, but now with interest targeting
  • Tests whether the winner also performs with specific interest audiences
  • Provides diversification alongside pure broad scaling

Destination C: Interest #2 Winners Ad Set

Winner Creative
      ↓
  TEST Campaign
      ↓
Interest #2 Winners
   (Ad Set)
  • Same creative with different interest targeting
  • Multiple interest ad sets possible depending on product/market

Complete Flow Summary

[NEW CREATIVE]
        │
        ▼
┌───────────────────┐
│   TEST Campaign   │
│                   │
│  Broad Pack [date]│ ◄── Starting point for all new creatives
│    (ABO budget)   │
└────────┬──────────┘
         │
         ▼
   [EVALUATION]
   Meets winner criteria?
         │
    ┌────┴────┐
    │         │
   NO        YES
    │         │
    ▼         ▼
  [KILL]   [WINNER]
    or        │
 [KEEP       │
 TESTING]    ├──────────────────────────────────────┐
             │                                      │
             ▼                                      ▼
┌────────────────────────┐          ┌───────────────────────────┐
│    SCALE Campaign      │          │      TEST Campaign        │
│         (ASC)          │          │                           │
│                        │          │  ┌─────────────────────┐  │
│  ┌──────────────────┐  │          │  │ Interest #1 Winners │  │
│  │   100% Broad     │  │          │  └─────────────────────┘  │
│  │   (all winners)  │  │          │  ┌─────────────────────┐  │
│  └──────────────────┘  │          │  │ Interest #2 Winners │  │
│                        │          │  └─────────────────────┘  │
│  Budget: 60-70%        │          │                           │
│  Goal: Maximum scale   │          │  Goal: Interest targeting │
└────────────────────────┘          │        with winners       │
                                    └───────────────────────────┘

Flow Rules Summary

RuleDescription
New creatives → Broad PackAll new creatives start in TEST campaign, Broad Pack ad set
Winner = multi-destinationWinners go to SCALE (ASC) AND Interest Winners ad sets
No new tests in SCALESCALE campaign contains only proven winners, no experiments
Interest Winners ≠ new testsInterest Winners ad sets contain graduated winners, not new creatives
Retarget/Retain = separate creativesThese campaigns use specific messaging, not the same test creatives

When Does a Creative NOT Advance?

A creative is paused (killed) and does not advance when:

  1. No conversions after 2-3x CPA spend → Pause immediately
  2. ROAS < 50% of target after significant spend → Pause
  3. CTR < 0.5% with €100+ spend → Pause or iterate
  4. Hook rate < 20% after 10,000+ impressions → Creative problem, iterate

Practical Example

Scenario: E-commerce brand, AOV €80, Target ROAS 4x, Target CPA €20

Week 1:
- Launch Broad Pack 2025-01-15 with Creative A, B, C, D
- Each creative gets equal budget (ABO)

Week 2 (after €60+ spend per creative):
- Creative A: €65 spend, 4 purchases, €400 revenue → ROAS 6.15x ✓
- Creative B: €70 spend, 2 purchases, €160 revenue → ROAS 2.29x (below threshold)
- Creative C: €55 spend, 0 purchases → KILL
- Creative D: €60 spend, 3 purchases, €280 revenue → ROAS 4.67x ✓

Actions:
- Creative A → Copy to SCALE + Interest #1 Winners + Interest #2 Winners
- Creative B → Keep testing (ROAS positive but below 3x threshold)
- Creative C → Pause (no conversions after 2.75x CPA)
- Creative D → Copy to SCALE + Interest #1 Winners + Interest #2 Winners

Campaign Settings per Tier

SCALE - Advantage+ Sales Campaign (ASC)

SettingValue
Campaign TypeAdvantage+ Sales Campaign
Budget TypeAdvantage Campaign Budget (formerly CBO)
Targeting100% Broad (Advantage+ Audience)
Existing Customer Cap0-10% (for pure acquisition)
CreativesOnly validated winners
Max ads150 per campaign, 50 per ad set

Scaling rules:

  • Increase budget by max 20-25% per 24-48 hours
  • Monitor frequency: >3.0 = creative fatigue risk
  • Add new winners, don't immediately pause old ones

TEST - Prospecting Campaign

SettingValue
Campaign TypeManual Sales Campaign
Budget TypeABO (Ad Set Budget Optimization)
TargetingBroad (Broad Packs) + Interest (Winners)
Structure3-5 creatives per Broad Pack ad set
Test durationMinimum 5-7 days

Broad Pack naming:

Ad Set: "Broad Pack [YYYY-MM-DD]"
├── Creative #1: [Format]_[Concept]_[Hook]_V1
├── Creative #2: [Format]_[Concept]_[Hook]_V2
├── Creative #3: [Format]_[Concept]_[Hook]_V3
└── Creative #4: [Format]_[Concept]_[Hook]_V4

RETARGET - Retargeting Campaign

AudienceWindowCreative Approach
FB/IG Engagers14 daysSocial proof, testimonials
Website Visitors30 daysProduct benefits, urgency
ATC/IC90 daysAbandoned cart, discount offers

Retargeting creative types:

  • Evergreen content (always relevant)
  • Objection handlers (FAQ, guarantees)
  • Sales/discount promotions
  • Low barrier offers (free shipping, samples)
  • Product-specific retargeting (Advantage+ Catalog Ads)

RETAIN - Retention Campaign

AudienceCreative Approach
All Time PurchasersCross-sell, new collections, VIP offers
180 Day PurchasersReplenishment, loyalty rewards, referral

Retention creative types:

  • Subscription/membership upsells
  • Bundle offers
  • Exclusive early access
  • Referral programs

Winner Criteria Framework

Step 1: Determine Client Parameters

Always ask the user first:

  1. What is the average order value (AOV)?
  2. What is the target CPA or ROAS?
  3. What is the business model? (e-commerce low/high AOV, lead gen)
  4. How much budget is available for testing?

Step 2: Calculate Winner Thresholds

Formula for E-commerce:

Minimum Spend = MAX(3 x Target CPA, 3 x AOV)
Minimum Conversions = Depends on AOV tier (see table)
ROAS Threshold = Depends on business model (see table)

Winner Criteria per Business Model:

Business ModelMin. ConversionsMin. SpendROAS Threshold
Low AOV (<€50)8-10 purchases3x CPA or ~€200-400>=4x ROAS
Medium AOV (€50-200)5-8 purchases3x CPA or ~€300-600>=3x ROAS
High AOV (>€200)3-5 purchases3x CPA or ~€500-1000>=2.5x ROAS
Lead Generation10-15 leads3x CPL or ~€200-500Lead Quality Score

Example calculation (Medium AOV):

AOV = €100
Target CPA = €25
Target ROAS = 4x (€100/€25)

Winner Criteria:
- Minimum spend: MAX(3 x €25, 3 x €100) = €300
- Minimum conversions: 5-8 purchases
- ROAS threshold: >=3x (conservative) or >=4x (target)

Step 3: Early Performance Signals (Video)

Evaluate after 24-48 hours and 10,000+ impressions:

MetricBenchmarkAction if Below
Hook Rate>=25% (target: 30-45%)Iterate first 3 seconds
Hold Rate>=40% (target: 60%+)Check pacing and narrative
CTR>=1% (target: 2%+)Test other hooks/angles
CPC<= account averageMonitor, don't pause yet

Step 4: Kill Criteria

Pause creative when:

  • No conversions after 2-3x target CPA spend
  • ROAS <50% of target after 5+ days
  • CTR <0.5% with significant spend (>€100)
  • Frequency >3.0 with declining performance
  • Hook rate <20% after 10,000+ impressions

Creative Velocity Guidelines

How many new creatives per month?

Monthly Ad SpendNew CreativesStructure
€0-10k1-2 per monthTest on-demand at fatigue
€10-25k3-4 per monthWeekly iteration cycle
€25-50k4-8 per month1-2 per week
€50-100k8-16 per month2-4 per week
€100k+15-50+ per monthDedicated creative team

Creative production breakdown:

  • ~50% iterations on proven concepts
  • ~30% new concepts within proven formats
  • ~20% experimental new formats/angles

Win rate expectations:

  • ~2% of creatives become true scale winners
  • ~10% show initial profitability
  • Iterations on winners often outperform originals

Budget Allocation Templates

Template A: Growth Focus (Recommended)

SCALE (ASC):     60%
TEST:            20%
RETARGET:        15%
RETAIN:           5%

Template B: Efficiency Focus

SCALE (ASC):     50%
TEST:            15%
RETARGET:        25%
RETAIN:          10%

Template C: New Account / Launch

SCALE (ASC):     40%  (or 0% until first winners)
TEST:            40%
RETARGET:        15%
RETAIN:           5%

Account Audit Checklist

Identifying Red Flags

Red FlagImpactAction
>5 active campaignsData fragmentationConsolidate to 4-tier model
Ad sets <50 conv/weekLearning LimitedMerge or increase budget
>50% budget in learningInefficiencyConsolidate, increase budgets
No dedicated test campaignNo pipelineImplement TEST tier
Retargeting >30% budgetOver-investmentReduce, focus on prospecting
Frequency >4 on scaleCreative fatigueRefresh creatives, add new

Audit Questions

  1. Structure: How many active campaigns? (Ideal: 2-4)
  2. Learning: How many ad sets in "Learning Limited"? (Ideal: <20%)
  3. Testing: Is there a dedicated test campaign? (Must: Yes)
  4. Winners: How are winners identified and scaled?
  5. Creative velocity: How many new creatives per month?
  6. Attribution: Is CAPI implemented with EMQ >=6?

MCP: Pull Account Structure for Audit

# Get active campaigns with delivery status and budget
meta_query(account_id="act_XXXXX", entity_type="campaigns", effective_status=["ACTIVE"], fields=["id","name","status","daily_budget","lifetime_budget","objective","buying_type"])

# Get ad sets for a campaign (learning status, audience info)
meta_query(account_id="act_XXXXX", entity_type="adsets", effective_status=["ACTIVE"], fields=["id","name","status","daily_budget","targeting","optimization_goal"])

Naming Convention

Campaign Level:

[Tier]_[Objective]_[BudgetType]_[Geo]

Example: SCALE_SALES_ACB_NL or TEST_SALES_ASB_EU

Ad Set Level:

[Tier]_[AudienceType]_[Date/Detail]

Example: TEST_BROAD_2025-01-15 or RT_WV30D_ALLPLC

Ad Level:

[Format]_[Concept]_[Hook]_[Version]

Example: VID_UGC_PROBLEM_V2 or IMG_LIFESTYLE_BENEFIT_V1

Code Legend

CodeMeaning
SCALEScaling campaign (ASC)
TESTTesting campaign
RTRetargeting
RETRetention
BROADAdvantage+ Audience / Broad
INTInterest targeting
LALLookalike Audience
WVWebsite Visitors
ENGEngagers (FB/IG)
ATCAdd to Cart
ICInitiate Checkout
PURCHPurchasers
VIDVideo
IMGStatic Image
CARCarousel
UGCUser Generated Content

Implementation Plan

Week 1: Audit & Setup

  1. Audit current account structure
  2. Define winner criteria with client
  3. Set up 4-tier campaign structure
  4. Implement naming conventions

Week 2: Testing Launch

  1. Launch first Broad Pack with 4-6 creatives
  2. Set up tracking and reporting dashboard
  3. Monitor early signals (hook rate, CTR)

Week 3-4: Optimize & Scale

  1. Evaluate creatives against winner criteria
  2. Graduate winners to SCALE/Interest
  3. Kill underperformers
  4. Launch new test batch

Ongoing: Maintain Velocity

  1. Weekly creative refresh cycle
  2. Monthly structure audit
  3. Quarterly winner criteria review

Output Template: Account Restructuring Plan

# Account Restructuring Plan: [Client Name]

## Current Situation
- Campaigns: [count]
- Ad Sets: [count]
- % in Learning Limited: [%]
- Primary issues: [list]

## Winner Criteria (Client-Specific)
- AOV: €[X]
- Target CPA: €[X]
- Target ROAS: [X]x
- Min. spend for winner: €[X]
- Min. conversions for winner: [X]

## Proposed 4-Tier Structure

### SCALE - ASC Campaign
- Budget: €[X]/day ([%] of total)
- Creatives: [current winners]

### TEST - Prospecting Campaign
- Budget: €[X]/day ([%] of total)
- First Broad Pack: [date]
- Creatives to test: [list]

### RETARGET Campaign
- Budget: €[X]/day ([%] of total)
- Audiences: [list with windows]

### RETAIN Campaign
- Budget: €[X]/day ([%] of total)
- Focus: [upsell/cross-sell strategy]

## Migration Timeline
- Week 1: [actions]
- Week 2: [actions]
- Week 3-4: [actions]

## Expected Impact
- CPA reduction: [%]
- ROAS improvement: [%]
- Learning Limited reduction: [%]

Files bundled with it

These load only when the skill asks for them, so they cost nothing until it runs.

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

Need help setting it up?

This page tells you what meta-campaign-structure-advisor does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.