agentleFS
gomarble-aigomarble-ai/marketing-agent/skills/meta-depth-of-analysis/SKILL.md

meta-depth-of-analysis

Skill15 starsChanged 39 days agoMITView on GitHub

Use for any Meta Ads audit, performance review, optimization request, strategy question, or funnel analysis. Covers 8 analytical dimensions: hierarchical drill-down, breakdowns, temporal trends, attribution rigor.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    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:

    Never make recommendations based solely on campaign-level or ad-set-level aggregates.

    line 15
  2. Distinctive:

    Do not apply universal KPIs.

    line 32
  3. Distinctive:

    Flag bid strategy issues: If a cost cap is set 3x above actual CPA, or tROAS target is 0.5x when achieving 3x — the algorithm isn't …

    line 48
  4. Distinctive:

    When this skill applies (audits, performance reviews, optimization), check at least 2 of these breakdown dimensions.

    line 52
  5. Distinctive:

    Statistical significance check: Do not make segment-level recommendations from <100 impressions or <3 conversions in a segment.

    line 67
  6. Distinctive:

    Never treat a date range as a single static number.

    line 71
  7. Distinctive:

    Do not analyze campaigns in isolation.

    line 87
  8. Distinctive:

    After answering the user's question, always surface 2-3 additional concerns or opportunities they did not ask about.

    line 117
  9. Distinctive:

    Do not take reported conversion numbers at face value.

    line 130

≈ “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 - Depth of Analysis

When to apply: Audits, performance reviews, optimization requests, strategy questions, funnel analysis. Skip for: Simple data fetches, campaign listings, metric lookups.

1. Hierarchical Drill-Down

Never make recommendations based solely on campaign-level or ad-set-level aggregates. Before concluding on any entity, drill one level deeper.

Required flow:

  • Asked about a campaign → check ad set performance within it
  • Asked about an ad set → check individual ad performance within it
  • Asked about ads → check creative-level patterns (same creative across multiple ads)

How:

facebook_get_adaccount_insights at level="adset" with campaign filter
facebook_get_adaccount_insights at level="ad" with adset filter

Identify which specific child entities are driving or dragging the parent's aggregate numbers. Quantify the contribution (e.g., "Ad Set X accounts for 72% of campaign spend but only 40% of conversions").

2. Objective-Contextual Evaluation

Do not apply universal KPIs. Evaluate each campaign against its actual objective.

Check first:

  • promoted_object.custom_event_type from ad set details
  • Campaign objective from facebook_get_campaign_details
  • Bid strategy (cost cap, bid cap, lowest cost)

Apply correct KPIs:

ObjectiveEvaluate onNOT on
Sales/PurchaseROAS, CPA, RevenueReach, CPM alone
Lead genCPL, lead volumeROAS
TrafficCPC, CTR, landing page viewsConversions
Video viewsCost per ThruPlay, hook rateROAS, CPA
AwarenessCPM, reach, frequencyCPA, ROAS

Flag bid strategy issues: If a cost cap is set 3x above actual CPA, or tROAS target is 0.5x when achieving 3x — the algorithm isn't actually constrained. Call this out.

3. Breakdown & Segmentation Analysis

When this skill applies (audits, performance reviews, optimization), check at least 2 of these breakdown dimensions. This overrides the general tool-fundamentals guidance to only use breakdowns when specifically requested — deep analysis requires segmentation.

Required breakdowns:

facebook_get_adaccount_insights with breakdowns="publisher_platform,platform_position"
facebook_get_adaccount_insights with breakdowns="device_platform"
facebook_get_adaccount_insights with breakdowns="age,gender"

What to look for:

  • Placement CPA differences (Feed vs Stories vs Reels vs Audience Network) — quantify the gap
  • Device performance divergence (mobile vs desktop)
  • Age/gender segments receiving delivery outside target — flag Advantage+ expansion leakage
  • Out-of-target segments consuming budget with poor conversion rates

Statistical significance check: Do not make segment-level recommendations from <100 impressions or <3 conversions in a segment. Call out low-confidence segments explicitly.

4. Temporal & Trend Analysis

Never treat a date range as a single static number. Always compare time windows.

Required comparisons:

facebook_get_adaccount_insights with time_increment="1" (daily data)

Analyze at minimum:

  • Last 7 days vs previous 7 days
  • Identify direction for each key metric: improving, declining, or stable
  • Flag inflection points: "CTR started declining on [date], 3 days before CPA started rising"

Connect the chain: Frequency rising → CTR falling → CPA rising → this is creative fatigue, not audience exhaustion. Correlate timing across metrics to establish cause, not just observe symptoms.

5. Cross-Entity Dependency Mapping

Do not analyze campaigns in isolation. Map the funnel.

Check:

  • Which campaigns are prospecting (broad/lookalike audiences)?
  • Which campaigns are retargeting (custom audiences, website visitors)?
  • Does cutting a prospecting campaign starve the retargeting audience pool?

How to detect:

  • facebook_get_adset_details → check targeting for custom audiences vs broad targeting
  • Prospecting: low frequency, lower CPM, lower CTR, broader audience
  • Retargeting: higher frequency, higher CPM, higher CTR, smaller audience

Flag dependencies: "Pausing Campaign X will reduce the Website Visitors 7d pool that feeds Campaign Y's retargeting. Expect retargeting CPA to rise within 1-2 weeks."

6. Platform Mechanic Awareness

Explain how Meta's mechanics affect the specific situation. Do not just mention them — explain the compound interaction.

Common compound effects to check:

SignalMechanicConsequence
Rising frequency + falling CTRCreative fatigueCPA will spike — need new creatives, not budget changes
CBO + one ad set getting 80% budgetCBO optimizationUsually working as intended — the algorithm concentrates budget on the best converter. Do not pause the dominant ad set just because smaller siblings look better on small samples; those often fail at scale. Only intervene if the dominant ad set itself is underperforming on ROAS/CPA.
Cost cap + rising CPMDelivery squeeze or audience shiftTwo possible causes — separate them before acting: (a) algorithm can't find conversions at the cap price (raise cap or loosen targeting); (b) original audience is saturated/exhausted and delivery is leaking to worse segments, pushing auction prices up. Check frequency, reach trend, and audience size to decide which.
Advantage+ audience expansion ONTargeting leakageCheck breakdown for out-of-target delivery eating budget
Insufficient spend on new ad setLearning phaseDo not make optimization decisions until the ad set has spent at least 2× AOV or 4× CPA, whichever is higher. The 50-conversion rule only applies to brand-new accounts with no prior conversion history.

7. Proactive Signal Detection

After answering the user's question, always surface 2-3 additional concerns or opportunities they did not ask about.

Scan for:

  • Any entity with frequency >3 and rising — creative fatigue incoming
  • Any ad set spending >30% of campaign budget with 0 conversions — budget waste
  • Any campaign where top ad has declining CTR over 7+ days — performance degradation ahead
  • Any retargeting audience pool that's shrinking (declining reach week over week)
  • Any metric that looks fine now but the trend predicts problems in 7-14 days

Format: "Outside your question, I noticed [signal] in [entity] — this suggests [predicted consequence] within [timeframe]."

8. Attribution Rigor

Do not take reported conversion numbers at face value.

Always check:

  • What attribution window is active? (1-day click, 7-day click, 1-day view?)
  • What share of conversions are view-through vs click-through?
  • Is the reported ROAS inflated by view-through on high-impression campaigns?

Flag these situations:

  • Retargeting campaign with very high ROAS + mostly view-through conversions → likely over-attributed
  • Two campaigns targeting the same audience → conversions may be double-counted
  • ASC/Advantage+ Shopping targeting existing customers → attributed ROAS includes conversions that would have happened organically

Frame as caveat: "Before scaling Campaign X based on its reported 4x ROAS, note that [specific attribution concern] means the true incremental ROAS is likely lower."

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.