meta-account-auditor

Account Auditor

Overview

This skill provides a systematic framework for performing complete Meta Ads account audits, including health scoring, issue identification, and prioritized recommendations for optimization.

Audit Framework

Audit Scope Overview

┌─────────────────────────────────────────────────────────────────┐
│  META ADS ACCOUNT AUDIT AREAS                                   │
│                                                                 │
│  1. TRACKING & DATA QUALITY                                     │
│     └── Pixel, CAPI, Events, Attribution                        │
│                                                                 │
│  2. ACCOUNT STRUCTURE                                           │
│     └── Campaigns, Ad Sets, Ads, Naming                         │
│                                                                 │
│  3. AUDIENCE STRATEGY                                           │
│     └── Custom, Lookalike, Targeting, Exclusions                │
│                                                                 │
│  4. CREATIVE PERFORMANCE                                        │
│     └── Formats, Fatigue, Testing, Diversity                    │
│                                                                 │
│  5. BUDGET & BIDDING                                            │
│     └── Allocation, Strategy, Efficiency                        │
│                                                                 │
│  6. PERFORMANCE METRICS                                         │
│     └── KPIs, Trends, Benchmarks                                │
│                                                                 │
│  7. COMPLIANCE & SETTINGS                                       │
│     └── Policies, Settings, Access                              │
└─────────────────────────────────────────────────────────────────┘

Complete Audit Checklist

Section 1: Tracking & Data Quality

TRACKING AUDIT
==============

Thresholds are rules of thumb, not Meta figures: check them against the account's own history.

□ PIXEL STATUS
├── Is the pixel installed and active?
├── Pixel ID correct on all relevant domains?
├── Test events in Events Manager
└── Score: [green OK / yellow Issues / red Critical]

□ CONVERSIONS API (CAPI)
├── Is CAPI implemented?
├── Event Match Quality score (goal: >6.0)
├── Deduplication correctly configured?
├── Which parameters are being sent?
└── Score: [green OK / yellow Issues / red Critical]

□ EVENT TRACKING
├── Which events are active?
├── Are events firing correctly? (test conversions)
├── Are custom events needed?
├── Value tracking for purchases?
├── Note: web events need no AEM setup or 8-event prioritization any more,
│   and domain verification is no longer required for event configuration
└── Score: [green OK / yellow Issues / red Critical]

□ ATTRIBUTION SETTING
├── Which setting does each ad set use? (Meta lists 1-day or 7-day click,
│   1-day view and 1-day engage-through; since 10 June 2025 Insights API results follow each ad set's setting)
├── Results under different settings are not directly comparable: use Compare attribution settings
├── Reports or rules still built on 7-day view or 28-day view windows, which the Insights API no longer returns (since 12 January 2026)?
└── Score: [green OK / yellow Issues / red Critical]

TRACKING SCORE: ___/40 points

Sources (checked 2026-10-10): Meta Business Help Center, "About Meta's Aggregated Event Measurement"; "About attribution models and attribution settings"; Marketing API reference, Ad Account Insights (action_attribution_windows: default means 7d_click and 1d_view).

Section 2: Account Structure

STRUCTURE AUDIT
===============

Ideal counts are rules of thumb, not Meta figures: check them against the account's own history.

□ CAMPAIGN ORGANIZATION
├── How many active campaigns? (rule of thumb: 3-8 for a typical account; the 4-tier model in campaign-structure-advisor uses 2-4)
├── Clear funnel segmentation? (TOFU/MOFU/BOFU)
├── Logical campaign objectives?
├── No duplicate campaigns?
└── Score: [green OK / yellow Issues / red Critical]

□ AD SET STRUCTURE
├── Number of ad sets per campaign (rule of thumb: 2-5)
├── Sufficient budget per ad set for learning? (Meta: about 50 results in the week after the last significant edit)
├── No overlapping audiences?
├── Logical audience segmentation?
└── Score: [green OK / yellow Issues / red Critical]

□ ADS ORGANIZATION
├── Per ad set: active ads, distinct concepts (look-alike ads count as one: Meta treats them as variations of the same creative) and the spend share of the top 3 ads. Descriptive only; Meta's diversification guidance gives no target number
├── Creative diversity present?
├── No duplicate ads?
├── Advantage+ creative and dynamic creative (two separate features) set up as intended?
└── Score: [green OK / yellow Issues / red Critical]

□ NAMING CONVENTIONS
├── Consistent naming system?
├── Campaign includes: objective, funnel, date?
├── Ad set includes: audience type?
├── Ad includes: creative type, version?
└── Score: [green OK / yellow Issues / red Critical]

STRUCTURE SCORE: ___/40 points

Section 3: Audience Strategy

AUDIENCE AUDIT
==============

Thresholds are rules of thumb, not Meta figures: check them against the account's own history.

□ CUSTOM AUDIENCES
├── Which custom audiences exist?
├── Are they up-to-date (recent data)?
├── Minimum size reached? (>1000)
├── Source quality assessment
└── Score: [green OK / yellow Issues / red Critical]

□ LOOKALIKE AUDIENCES
├── Which LALs are active?
├── Source audience quality?
├── Percentage selection appropriate?
├── Are value-based LALs being used?
├── Note: with Advantage+ audience on, Meta can expand beyond inputs such as lookalikes;
│   location, minimum age, language and custom audience exclusions stay fixed
└── Score: [green OK / yellow Issues / red Critical]

□ INTEREST/BEHAVIOR TARGETING
├── Relevance of selected interests?
├── Audience size not too broad/narrow?
├── Advantage+ audience A/B tested on prospecting? (Meta recommends testing it for almost all campaign types, not for retargeting)
├── Testing of new interests?
└── Score: [green OK / yellow Issues / red Critical]

□ EXCLUSIONS
├── Are buyers excluded where needed?
├── Funnel-based exclusions active?
├── Audience overlap minimized?
├── No conflicting exclusions?
└── Score: [green OK / yellow Issues / red Critical]

AUDIENCE SCORE: ___/40 points

Sources (checked 2026-10-10): Meta Business Help Center, "About Advantage+ audience"; Marketing API, "Advantage+ audience".

Section 4: Creative Performance

CREATIVE AUDIT
==============

□ FORMAT DIVERSITY
├── Mix of video, static, carousel?
├── Different aspect ratios (1:1, 4:5, 9:16)?
├── UGC vs produced content balance?
├── Stories/Reels-specific creative?
└── Score: [green OK / yellow Issues / red Critical]

□ CREATIVE HEALTH
├── Average age of active creatives?
├── CTR trend (rising/falling)?
├── Frequency per creative?
├── Creative fatigue signals?
└── Score: [green OK / yellow Issues / red Critical]

□ MESSAGING & COPY
├── Clear value proposition?
├── Strong hooks (first 3 sec)?
├── Variation in copy angles?
├── CTAs clear and consistent?
└── Score: [green OK / yellow Issues / red Critical]

□ TESTING CADENCE
├── Are new creatives being tested?
├── A/B testing active?
├── Clear test hypotheses?
├── Learnings documented?
└── Score: [green OK / yellow Issues / red Critical]

CREATIVE SCORE: ___/40 points

Section 5: Budget & Bidding

BUDGET AUDIT
============

□ BUDGET ALLOCATION
├── Budget distribution across funnel?
├── No ad sets under minimum budget?
├── Budget aligned with goals?
├── Pacing correct (daily vs lifetime)?
└── Score: [green OK / yellow Issues / red Critical]

□ BID STRATEGY
├── Which bid strategies in use?
├── Appropriate for campaign objectives?
├── Cost caps realistic?
├── ROAS targets achievable?
└── Score: [green OK / yellow Issues / red Critical]

□ BUDGET EFFICIENCY
├── Spend vs budget ratio?
├── Are budgets fully spending?
├── ROI per campaign/funnel stage?
├── Identification of budget waste?
└── Score: [green OK / yellow Issues / red Critical]

□ SCALING POTENTIAL
├── Winning ad sets with room to scale?
├── Headroom in audiences?
├── Budget increase history successful?
└── Score: [green OK / yellow Issues / red Critical]

BUDGET SCORE: ___/40 points

Section 6: Performance Review

PERFORMANCE AUDIT
=================

□ KEY METRICS (vs benchmarks)
├── CPA: $___ (benchmark: $___)
├── ROAS: ___ (benchmark: ___)
├── CTR: ___% (benchmark: ___%)
├── CPM: $___ (benchmark: $___)
└── Score: [green OK / yellow Issues / red Critical]

□ TREND ANALYSIS (Last 30 days)
├── CPA trend: [up / flat / down]
├── ROAS trend: [up / flat / down]
├── Volume trend: [up / flat / down]
├── Efficiency trend: [up / flat / down]
└── Score: [green OK / yellow Issues / red Critical]

□ FUNNEL METRICS
├── TOFU: Reach, CPM, Video Views
├── MOFU: CTR, CPC, Engagement
├── BOFU: CPA, ROAS, Conv Rate
└── Score: [green OK / yellow Issues / red Critical]

□ LEARNING PHASE STATUS
├── % ad sets in learning?
├── Average time in learning?
├── Learning phase success rate?
└── Score: [green OK / yellow Issues / red Critical]

PERFORMANCE SCORE: ___/40 points

Section 7: Unused Feature Detection

UNUSED FEATURES AUDIT
=====================
This section identifies money left on the table - features that are
available but not being used, representing optimization opportunities.
Thresholds are rules of thumb, not Meta figures: check them against the account's own history.

□ ADVANTAGE+ FEATURES (Underutilized?)
├── Advantage+ Sales Campaigns
│   └── Sales objective without an Advantage+ sales campaign? (Meta recommends
│       Advantage+ campaigns for sales, leads and app promotion objectives)
├── Advantage+ Placements
│   └── Manual placements only? Missing reach opportunities
├── Advantage Detailed Targeting
│   └── Disabled? AI audience expansion not working
├── Advantage+ Creative
│   └── Not using? Missing creative optimization
└── Score: [green Using / yellow Partially / red Not Using]

□ CONVERSIONS API (CAPI) - Enabled but Low Volume?
├── Conversions API event coverage below 75% of Pixel events? (Meta's goal)
│   └── Missing: Better attribution
├── Only PageView sent? Missing Add-to-Cart, Purchase?
│   └── Missing: Conversion signal strength
├── Event Match Quality <6.0?
│   └── Missing: Full CAPI value (poor matching)
└── Score: [green OK / yellow Partial / red Not Optimized]

□ AUDIENCE ASSETS (Created but Not Used?)
├── Custom Audiences uploaded but not in active ad sets?
│   └── Wasted asset: Customer data not being leveraged
├── Lookalike Audiences created but inactive?
│   └── Wasted asset: Expansion opportunities unused
├── Value-based Lookalikes available but not used?
│   └── Missing: Higher-value customer targeting
├── Website remarketing lists exist but no retargeting campaigns?
│   └── Wasted: Warm audiences not being targeted
└── Score: [green Using All / yellow Partially / red Many Unused]

□ CATALOG & SHOPPING FEATURES (E-commerce)
├── Catalog connected but no Advantage+ catalog ads (formerly dynamic ads)?
│   └── Missing: Personalized product recommendations
├── Advantage+ creative for catalog not used?
│   └── Missing: Automated creative optimization
├── Collection Ads not tested?
│   └── Missing: Mobile shopping experience
├── Collaborative Ads available (retailer) but not used?
│   └── Missing: Partner reach
└── Score: [green OK / yellow Partial / red Underutilized]

□ AUTOMATION & RULES (Not Configured?)
├── No automated rules for budget/bid management?
│   └── Missing: Proactive management, less manual work
├── No Advantage Campaign Budget enabled?
│   └── Missing: Automated budget allocation across ad sets
├── No cost caps or bid caps when needed?
│   └── Risk: No control over cost per result (budgets cap spend)
├── No scheduled ad set rules (day/time)?
│   └── Missing: Performance optimization by schedule
└── Score: [green Using / yellow Basic / red None]

□ MEASUREMENT & ATTRIBUTION (Not Set Up?)
├── Conversion Lift studies available but not run?
│   └── Missing: True incrementality data
├── Brand Lift studies available but not used?
│   └── Missing: Brand impact measurement
├── A/B testing (Experiments) not used?
│   └── Missing: Statistical testing framework
├── Compare attribution settings not reviewed?
│   └── Missing: Understanding of attribution impact
└── Score: [green Using / yellow Basic / red None]

UNUSED FEATURES SCORE: ___/60 points

Sources (checked 2026-10-10): Meta Business Help Center, "About Advantage+ audience" (Meta's recommendation of Advantage+ campaigns); Marketing API, "Advantage+ Campaign Experience for Sales, App, and Leads" (Advantage+ sales campaigns replace Advantage+ shopping campaigns, which can no longer be created through the API); Conversions API, "Dataset Quality API" (event coverage is the 7-day average share of Pixel events that the Conversions API also sends with matching deduplication keys; Meta gives 75% as the goal).

Health Score Calculator

Score Interpretation

ACCOUNT HEALTH SCORE
====================

SECTION               │ MAX POINTS │ YOUR SCORE
─────────────────────┼────────────┼───────────
Tracking & Data      │     40     │    ___
Account Structure    │     40     │    ___
Audience Strategy    │     40     │    ___
Creative Performance │     40     │    ___
Budget & Bidding     │     40     │    ___
Performance Metrics  │     40     │    ___
Unused Features      │     60     │    ___
─────────────────────┼────────────┼───────────
TOTAL                │    300     │    ___

SCORE INTERPRETATION (our scoring model, not a Meta measure):
├── 250-300: Excellent - Minor optimizations
├── 200-249: Good - Several improvements needed
├── 150-199: Fair - Significant work required
└── <150:    Poor - Major restructuring needed

Issue Prioritization Matrix

Categorize found issues:

ISSUE PRIORITIZATION
====================

Thresholds are rules of thumb, not Meta figures: check them against the account's own history.

CRITICAL (Fix within 24 hours):
├── Tracking not working
├── No conversions being tracked
├── Budget completely stopped
├── Policy violations
└── Major data discrepancies

HIGH (Fix within 1 week):
├── Poor EMQ score (<5)
├── High audience overlap (>50%)
├── Creative fatigue (CTR <0.5%)
├── CPA >50% above target
└── Learning phase failures

MEDIUM (Fix within 2 weeks):
├── Naming conventions inconsistent
├── Ad set consolidation needed
├── Audience refresh needed
├── Budget reallocation needed
└── Missing exclusions

LOW (Fix within 1 month):
├── Nice-to-have optimizations
├── Testing opportunities
├── Documentation updates
└── Minor structural changes

Audit Report Template

When the user asks for an account audit:

ACCOUNT AUDIT REPORT
=====================

Audit Date: [DATE]
Account: [ACCOUNT NAME]
Auditor: Claude AI

EXECUTIVE SUMMARY
────────────────────
Account Health Score: [X]/300 ([STATUS])

Top 3 Strengths:
1. [Strength]
2. [Strength]
3. [Strength]

Top 3 Areas for Improvement:
1. [Area]
2. [Area]
3. [Area]

DETAILED FINDINGS
─────────────────────────────

1. TRACKING & DATA
   Status: [green/yellow/red]
   Issues: [List]
   Recommendations: [List]

2. ACCOUNT STRUCTURE
   Status: [green/yellow/red]
   Issues: [List]
   Recommendations: [List]

[...etc for all sections...]

ACTION PLAN (Prioritized)
────────────────────────────

WEEK 1 (Critical):
□ [Action 1]
□ [Action 2]

WEEK 2-3 (High):
□ [Action 3]
□ [Action 4]

MONTH 1 (Medium):
□ [Action 5]
□ [Action 6]

EXPECTED IMPACT
───────────────
Estimate only from this account's own data and state the assumptions;
otherwise write "not estimated":
├── CPA: [current value] → [estimate, if supported]
├── ROAS: [current value] → [estimate, if supported]
└── Efficiency: [current value] → [estimate, if supported]

Quick Audit Checklist

For a fast health check:

QUICK AUDIT (15 minutes)
========================

Thresholds are rules of thumb, not Meta figures: check them against the account's own history.

□ Pixel active? (Events Manager → Test Events)
□ Conversions last 7 days? (yes/no/how many)
□ Learning phase issues? (% in learning)
□ CPA vs target? (under/on/over)
□ Frequency <3? (yes/no)
□ CTR >1%? (yes/no)
□ Genuinely different concepts in each ad set, not only look-alike variations? (yes/no)
□ Budget fully spending? (yes/no)

QUICK SCORE: ___/8
└── 7-8: Healthy account
└── 5-6: Attention needed
└── <5: In-depth audit needed

MCP Tool Integration

Running an Audit with Live Data

Use these MCP calls to pull actual account data before scoring each section:

1. Campaign structure overview:
→ meta_query(account_id="...", entity_type="campaigns", effective_status=["ACTIVE"], fields=["id","name","objective","status","daily_budget","lifetime_budget"])

2. Ad set performance + delivery:
→ meta_get_insights(account_id="...", level="adset", date_preset="last_7d", fields=["cost_per_action_type","spend","frequency","impressions"])

3. Ad-level creative health (ranks by spend within the first 100 ad rows of the period; 10 ads by default, max 25):
→ meta_get_creatives(account_id="...", scope="account", date_preset="last_7d")

4. Top/bottom performers (last 30 days):
→ meta_get_insights(account_id="...", level="campaign", date_preset="last_30d", fields=["spend","actions","cost_per_action_type","impressions"])

5. Learning phase status (active ad sets only; not returned for ad sets with dynamic creative, and not every ad account gets it):
→ meta_query(account_id="...", entity_type="adsets", effective_status=["ACTIVE"], fields=["id","name","learning_stage_info"], format="full")

6. Pixel activity:
→ meta_query(account_id="...", entity_type="adspixels")   (last_fired_time per pixel)
→ meta_query(account_id="...", entity_type="dataset_quality", entity_id="<pixel_id>")   (EMQ, match keys and dedup per web event; status "unavailable" means no access, not bad tracking)

meta_get_insights returns one page of rows: while a response has paging.next, fetch the next page with
meta_query(account_id="...", entity_type="adsets", pagination_url=<paging.next>) and add its rows.

Feed results directly into the audit sections above to score each area objectively.

Example Interaction

User: "I just took over a new account. Can you help me with an audit?"

Claude response:

"I'll help you with a systematic account audit. Let me pull the live data first:

meta_get_insights(account_id="...", level="campaign", date_preset="last_7d", fields=["spend","cost_per_action_type","frequency","impressions"])

meta_get_insights(account_id="...", level="adset", date_preset="last_7d", fields=["cost_per_action_type","spend","frequency"])

While that runs, I also need a few things:

---

STEP 1: Gather Basic Info

Can you tell me the following:

1. What is the industry/vertical?
2. What is the monthly ad budget?
3. What are the current KPI targets (CPA/ROAS)?
4. How long has the account been running?

---

STEP 2: Key Things I'll Check From the Data

Thresholds below are rules of thumb, not Meta figures: check them against the account's own history.

Tracking Check:

• Pixel active (last_fired_time); EMQ for the Purchase event from meta_query(account_id, "dataset_quality", entity_id=<pixel_id>) (target: 6.0+)
• Attribution setting per ad set (Meta lists 1-day or 7-day click, 1-day view and 1-day engage-through)

Structure Check:

• Number of active campaigns (rule of thumb: 3-8; 2-4 in the 4-tier model)
• Ad sets per campaign (rule of thumb: 2-5)
• Creative count and diversity

Performance Check (Last 7 days):

• CPA vs target
• ROAS vs target
• Learning phase % (should be <20% of budget; status from learning_stage_info)

---

Share the account ID and I'll generate a complete audit with health score and action plan!"