agentleFS
ghazanfar111212ghazanfar111212/claude-skills-performance-marketing/skills/meta-ads-expert/SKILL.md

meta-ads-expert

Skill0 starsChanged 38 days agoMITView on GitHub

Expert-level Meta Ads (Facebook/Instagram) skill. Use whenever the user mentions Meta Ads, Facebook Ads, Instagram Ads, FB Ads Manager, Business Manager, Advantage+ Shopping, ASC, CBO/ABO, Lookalike audiences, Custom Audiences, Conversions API (CAPI), iOS 14.5 attribution, aggregated event measurement, pixel setup, catalog ads, DPA, creative testing, 1-day click attribution, or any paid social campaign on Meta platforms. Also trigger for "run Facebook ads," "Instagram lead gen," "creative iteration," "UGC ad scripts," "hook testing," or auditing a Meta ad account. Do NOT use for organic social, TikTok ads, LinkedIn ads, or Google Ads. Covers account structure, campaign objectives, bidding, audience strategy, creative frameworks, CAPI setup, attribution, scaling playbooks, and troubleshooting.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

13 rules. About 0 are common, 13 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈13

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Test 3-5 hooks against same body/CTA to isolate variable

    line 62
  2. Distinctive:

    Verify domain in Business Manager

    line 83
  3. Distinctive:

    Default: 7-day click / 1-day view

    line 88
  4. Distinctive:

    Don't edit ad sets mid-learning (restarts the phase)

    line 96
  5. Distinctive:

    Cost Cap: When you have reliable CPA data and want to scale predictably

    line 105
  6. Distinctive:

    Don't 2x overnight - kills learning

    line 113
  7. Distinctive:

    Always report attribution window used

    line 136
  8. Distinctive:

    Don't use Traffic objective hoping to get conversions

    line 141
  9. Distinctive:

    Don't stack multiple Lookalikes in one ad set (they overlap)

    line 142
  10. Distinctive:

    Don't run <$20/day ad sets on lead gen - won't exit learning

    line 143
  11. Distinctive:

    Don't optimize for landing page views on a Sales objective campaign

    line 144
  12. Distinctive:

    Don't edit live ads mid-learning unless killing them

    line 145
  13. Distinctive:

    Don't trust 1-day view attribution for paid social (inflates results)

    line 146

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Meta Ads Expert

When to Use This Skill

  • Building or restructuring a Meta Ads account
  • Planning campaign architecture (CBO vs ABO, Advantage+ vs manual)
  • Creative strategy, hook testing, UGC scripting
  • Troubleshooting delivery, frequency, CPM spikes, or low CTR
  • Setting up Conversions API and resolving iOS attribution gaps
  • Scaling winning campaigns without breaking learning phase
  • Auditing an existing account

Core Account Architecture

The 3-Tier Hierarchy

  • Campaign: objective + budget (if CBO) + bid strategy
  • Ad Set: audience + placements + budget (if ABO) + optimization event
  • Ad: creative + copy + CTA + destination

Campaign Objectives (2024+ ODAX framework)

  • Sales - purchase/lead conversions (most common for DTC & services)
  • Leads - instant forms or website leads
  • Traffic - rarely worth using; only for top-funnel content distribution
  • Engagement - retargeting warm audiences, message conversions
  • Awareness - reach/brand lift (mid-to-enterprise only)
  • App Promotion - app installs/events

Account Structure Playbooks

For DTC Ecommerce

  • 1 Advantage+ Shopping Campaign (ASC) - 80% of budget
  • 1 Broad Targeting Campaign (manual) - 15%
  • 1 Retargeting Campaign (180-day window) - 5%
  • Creative refresh cadence: 2-4 new ads/week into ASC

For Lead Gen (Services, Dental, Contractors)

  • 1 Conversion Campaign optimizing for lead event (not form fills)
  • Broad audience + detailed targeting expansion ON
  • Location-specific ad sets if serving multiple markets
  • Instant Form + Website Lead split test

For High-Ticket B2B

  • Lead campaign with enriched Instant Form (qualifying questions)
  • Lookalike 1% of past closed-won clients
  • CAPI with offline conversion uploads critical
  • Longer attribution window (7-day click/1-day view)

Creative Framework

The Hook-Problem-Solution-Proof-CTA Framework

  1. Hook (first 3 seconds): Pattern interrupt, curiosity, bold claim, or relatability trigger
  2. Problem: Agitate pain point the audience feels
  3. Solution: Introduce product as the fix
  4. Proof: Testimonial, before/after, data, social proof
  5. CTA: Clear action + urgency

Creative Testing Methodology (Hook Testing)

  • Test 3-5 hooks against same body/CTA to isolate variable
  • Minimum $50-100 spend per hook before calling it
  • CTR (outbound) and Hook Rate (3s video views / impressions) are your leading indicators
  • Scale winners, kill losers at statistical significance

Creative Diversity Requirements

Per ad set, run at minimum:

  • 1 UGC talking-head video
  • 1 product demo or static
  • 1 testimonial/review format
  • 1 problem-solution static or carousel

iOS 14.5+ Attribution Strategy

Conversions API (CAPI) - Non-Negotiable

  • Deploy server-side via GTM Server, Stape, or native Shopify/WP integration
  • Deduplicate with Pixel events using event_id
  • Match quality score target: 7+ out of 10
  • Send these events minimum: PageView, ViewContent, AddToCart, InitiateCheckout, Purchase, Lead

Aggregated Event Measurement (AEM)

  • Verify domain in Business Manager
  • Prioritize 8 events per domain (Purchase/Lead always in slot 1)
  • Value optimization requires value parameter + slot reservation

Attribution Windows

  • Default: 7-day click / 1-day view
  • For considered purchases ($500+): stick with 7-day click
  • For impulse DTC: 1-day click can speed up learning

Budget & Bidding

Learning Phase Rules

  • 50 optimization events per ad set per week to exit learning
  • Don't edit ad sets mid-learning (restarts the phase)
  • Starting budget = 3-5x target CPA minimum

CBO vs ABO Decision Tree

  • CBO: 3+ ad sets, similar audience sizes, proven creatives
  • ABO: Testing new audiences, uneven ad set sizes, constrained budgets

Bid Strategies

  • Highest Volume (default): Use 80% of the time
  • Cost Cap: When you have reliable CPA data and want to scale predictably
  • Bid Cap: Advanced only - when you know exact auction dynamics
  • ROAS Goal: Ecommerce with consistent AOV and value optimization

Scaling Playbooks

Vertical Scaling (same campaign)

  • Increase budget by 20-30% every 3-4 days if CPA holds
  • Don't 2x overnight - kills learning

Horizontal Scaling

  • Duplicate winning ad set to new audience (Lookalike variations)
  • Duplicate to new placement (Reels-only, Stories-only)
  • Geographic expansion (new markets/countries)

Creative Scaling

  • Take winning ad → 5 variations (new hook, new format, new CTA)
  • Iterate fast; winning creative has a shelf life of 2-4 weeks

Troubleshooting Cheat Sheet

SymptomLikely CauseFix
High CPM, low CTRCreative fatigueRefresh creative, expand audience
Low deliveryAudience too narrow or bid too lowBroaden audience, raise bid/budget
Learning Limited<50 events/weekRaise budget or consolidate ad sets
CAPA worse than last monthCreative/offer/market shiftAudit creative performance, check seasonality
Pixel/CAPI mismatchDeduplication brokenVerify event_id pass-through

Reporting Standards

  • Primary KPIs: CPA, ROAS, Frequency, CTR (outbound), Hook Rate
  • Always report attribution window used
  • Separate platform (Meta reported) vs source of truth (GA4/Shopify)
  • Triangulate with post-purchase surveys for brand awareness spend

Anti-Patterns

  • Don't use Traffic objective hoping to get conversions
  • Don't stack multiple Lookalikes in one ad set (they overlap)
  • Don't run <$20/day ad sets on lead gen - won't exit learning
  • Don't optimize for landing page views on a Sales objective campaign
  • Don't edit live ads mid-learning unless killing them
  • Don't trust 1-day view attribution for paid social (inflates results)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.