account-structure-review

Account Structure Review — Google + Meta

Expert-level guidance for evaluating paid media account structure against goals, budget, and conversion volume thresholds. Identifies structural anti-patterns that fragment data, starve budgets, or block algorithms from learning.

Install

git clone https://github.com/thatrebeccarae/claude-marketing.git && cp -r claude-marketing/skills/account-structure-review ~/.claude/skills/

Core Capabilities

Structural Anti-Pattern Detection

• Over-segmentation: Too many campaigns/ad sets splitting data and budget below learning thresholds
• Under-segmentation: Everything in one campaign, can't isolate performance by intent or funnel stage
• Budget fragmentation: Daily budgets too low to exit learning phase or sustain delivery
• Targeting overlap: Multiple campaigns competing in same auctions, bidding against yourself
• Objective mismatch: Using Traffic when you need Conversions, mixing lead gen and ecommerce goals in one campaign

Google-Specific Structure Auditing

• Campaign type separation (Search ≠ Display ≠ Shopping ≠ PMax)
• Ad group theme coherence (SKAGs are dead, STAGs with 5-15 keywords)
• Match type strategy alignment per campaign intent
• PMax cannibalizing branded Search
• Shared budgets causing cross-campaign interference
• Search Partners eating budget without conversions
• Automated bid strategies applied to low-volume campaigns

Meta-Specific Structure Auditing

• CBO vs ABO alignment with goals and budget levels
• Advantage+ Shopping adoption readiness
• Prospecting vs retargeting separation
• Funnel exclusions (retargeting excludes purchasers, prospecting excludes retargeting audiences)
• Advantage+ Audience stability (broad targeting with signal-based constraints)
• Campaign duplication testing without isolation
• Multiple ad sets in prospecting CBO campaigns (common waste vector)

Consolidation Decision Framework

• How to merge campaigns without losing historical data
• Campaign experiments for migration testing (Google) or 50/50 A/B tests (Meta)
• Incremental changes vs wholesale restructure
• Sequencing: What to consolidate first, what to defer

Structure Health Criteria

Google Ads: Minimum Conversion Volume Thresholds

| Bid Strategy | Minimum Monthly Conversions | Ideal Monthly Conversions | Warning Signs | |--------------|----------------------------|--------------------------|---------------| | Manual CPC | N/A (no learning required) | N/A | Using manual in 2026 means you're leaving performance on the table | | Maximize Clicks | 0 (but why?) | N/A | You have conversion tracking — use it | | Maximize Conversions | 15/month | 30+/month | <15/month = erratic, algorithm thrashing | | Target CPA | 30/month | 50+/month | <30/month = CPA target won't be hit reliably | | Target ROAS | 50/month | 100+/month | <50/month = ROAS variance too high, poor signal |

Learning Phase: Campaigns need ~50 conversions in 30 days to exit learning. Campaigns with <30 conversions/month perpetually re-enter learning with every creative refresh or bid change.

Meta: Ad Set-Level Conversion Volume Thresholds

| Optimization Goal | Minimum Weekly Conversions | Ideal Weekly Conversions | Warning Signs | |-------------------|----------------------------|--------------------------|---------------| | Landing Page Views | N/A (top funnel) | N/A | Using this when you have purchase tracking = waste | | Link Clicks | N/A (top funnel) | N/A | Same as above | | Leads (Lead Gen) | 10/week per ad set | 20+/week | <10/week = unstable CPL | | Add to Cart | 15/week per ad set | 30+/week | <15/week = algorithm can't optimize reliably | | Initiate Checkout | 20/week per ad set | 40+/week | <20/week = CPA variance too high | | Purchase | 50/week per ad set | 100+/week | <50/week = learning limited, poor delivery | | Purchase (Advantage+) | 50/week per campaign | 150+/week | Advantage+ needs high volume, consolidate aggressively |

Learning Phase: Ad sets need ~50 conversions/week to exit learning. Ad sets with <50/week perpetually re-enter learning with every edit.

Budget Fragmentation Thresholds

| Platform | Minimum Daily Budget | Ideal Daily Budget | Why | |----------|---------------------|-------------------|-----| | Google Search (tCPA) | $150/day | $300+/day | 10x target CPA minimum for stable delivery | | Google PMax | $100/day | $300+/day | PMax needs volume to test assets across channels | | Google Display | $50/day | $100+/day | Lower conversion rate = need more spend to gather signal | | Meta CBO (Conversions) | $100/day | $300+/day | CBO needs budget to test ad sets before allocating | | Meta ABO (Conversions) | $20/day per ad set | $50+/day per ad set | Ad sets under $20/day starve before learning completes | | Meta Advantage+ | $200/day | $500+/day | Consolidated campaigns need higher budgets to sustain delivery |

Critical Rule: If your campaign can't sustain the minimum daily budget for 30 consecutive days without pausing, your account is over-segmented.

Targeting Overlap Indicators (Meta)

| Overlap % | Status | Action Required | |-----------|--------|-----------------| | 0-10% | Healthy | No action | | 11-25% | Moderate | Monitor, consider consolidation if both ad sets underperform | | 26-50% | High | Consolidate or add exclusions immediately | | 51%+ | Critical | You're bidding against yourself, merge now |

Check overlap in Meta Ads Manager > Tools > Audience Overlap. Run this monthly for all active prospecting campaigns.

Google: Keyword Overlap Indicators

| Scenario | Status | Action Required | |----------|--------|-----------------| | Same keyword, different match types, same campaign | Healthy | This is fine, broad/phrase/exact in same ad group is valid | | Same keyword, same match type, different campaigns | High Risk | Unless separated by geo/device/schedule, you're competing with yourself | | Broad keyword in Campaign A matching exact keyword in Campaign B | Critical | Broad will steal traffic from exact unless you add exact as negative in Campaign A | | PMax active + Brand Search campaign | Critical | PMax WILL cannibalize branded search. Add brand terms as negatives in PMax or pause PMax. |

Run Search Terms Report monthly and cross-reference against all campaigns to find cannibalization.

Platform-Specific Structure Best Practices

Google Ads: Campaign Hierarchy

Search Campaigns (Hierarchy: Campaign > Ad Group > Keywords > Ads)

• One theme per campaign: Brand, Competitor, Category, Product Line, Intent Stage
• Ad groups: 5-15 keywords per ad group, tightly themed (STAGs, not SKAGs)
• Match types: Use Phrase and Broad in most campaigns. Exact only for ultra-high-intent/high-value terms.
• Negatives: Aggressive negative keyword lists at campaign level to prevent overlap
• Structure example:
• Campaign: Brand Search
• Ad Group: Brand Exact (5 exact match brand terms)
• Ad Group: Brand Misspellings (10 phrase match variants)
• Campaign: Category - Running Shoes
• Ad Group: Trail Running (8 phrase match keywords)
• Ad Group: Road Running (10 phrase match keywords)
• Ad Group: Racing Flats (5 phrase match keywords)

Shopping Campaigns

• Standard Shopping: One campaign per product category OR one campaign with ad groups by product type
• PMax: One campaign for entire catalog OR separate campaigns by margin/goal (e.g., high-margin PMax vs volume PMax)
• Do NOT run PMax + Standard Shopping simultaneously — PMax will dominate, Standard Shopping becomes a reporting shell

Performance Max (PMax)

• When to use: E-commerce with feed, lead gen with dynamic assets, omnichannel goals
• Structure: 1-3 campaigns max. PMax is designed to consolidate. Multiple PMax campaigns only if goals differ (ROAS target, audience strategy, geo)
• Asset groups: 3-5 per campaign, themed by product category or customer segment
• Critical negative keywords: Add brand te