google-ads-pmax-asset-group-optimizer

Performance Max Asset Group Optimizer

Complete guide for optimizing Performance Max asset groups — from structure and audience signals to creative testing and feed optimization.

Asset group count decision guide in references/decision-trees.md.

MCP Tool Integration

STEP 1: PMax campaign performance overview
google_ads_run_gaql(query="
  SELECT campaign.name, campaign.status,
    metrics.cost_micros, metrics.conversions,
    metrics.cost_per_conversion, metrics.conversions_value
  FROM campaign
  WHERE campaign.advertising_channel_type = 'PERFORMANCE_MAX'
    AND campaign.status = 'ENABLED'
    AND segments.date DURING LAST_30_DAYS
  ORDER BY metrics.cost_micros DESC
")

STEP 2: Asset group performance
google_ads_run_gaql(query="
  SELECT campaign.name, asset_group.name, asset_group.status,
    metrics.impressions, metrics.clicks, metrics.conversions,
    metrics.cost_micros, metrics.cost_per_conversion
  FROM asset_group
  WHERE campaign.advertising_channel_type = 'PERFORMANCE_MAX'
    AND segments.date DURING LAST_30_DAYS
  ORDER BY metrics.cost_micros DESC
")

STEP 3: Search terms (v21+)
google_ads_run_gaql(query="
  SELECT campaign.name, search_term_view.search_term,
    metrics.impressions, metrics.clicks,
    metrics.conversions, metrics.cost_micros
  FROM search_term_view
  WHERE campaign.advertising_channel_type = 'PERFORMANCE_MAX'
    AND segments.date DURING LAST_30_DAYS
  ORDER BY metrics.cost_micros DESC
  LIMIT 100
")

STEP 4: Ad network type breakdown (v22+)
google_ads_run_gaql(query="
  SELECT campaign.name, segments.ad_network_type,
    metrics.impressions, metrics.clicks,
    metrics.conversions, metrics.cost_micros
  FROM campaign
  WHERE campaign.advertising_channel_type = 'PERFORMANCE_MAX'
    AND segments.date DURING LAST_30_DAYS
")

Asset Group Structure

Structuring Options

Option	Best for	Example
Category-based	Most common	Women's / Men's / Accessories
Margin-based	Profit focus	High / Standard / Clearance
Funnel-based	Customer journey	Prospecting / Retargeting / Cross-sell
Performance-based	Data-driven	Top Performers / Testing / Long Tail
Hybrid (recommended)	Best of all	Bestsellers + Category A/B/C + Sale

Asset Requirements

Asset Type	Min	Recommended	Max	Specs
Headlines	3	5-15	15	Max 30 chars
Long Headlines	1	3-5	5	Max 90 chars
Descriptions	2	4-5	5	Max 90 chars
Images	2	6-10	20	1200x628, 1200x1200, 960x1200
Logos	1	2	5	1200x1200 (square), 1200x300 (landscape)
Videos	0	1-3	5	15-30 sec, 16:9 + 1:1 + 9:16

No video upload = Google auto-creates video from images (often less effective).

Detailed asset variation strategy (headline/image/video types) in references/detailed-reference.md.

Asset Performance Ratings

Per-asset ratings (Best/Good/Low/Pending) removed from API in v22 — check in Google Ads UI only (Campaign → Asset Groups → Asset Performance).

Optimization cycle:

• Week 1-2: Learning phase — upload all assets, make no changes
• Week 3-4: First analysis — identify "Low" performers, note trends
• Week 5-6: First optimization — replace consistently "Low" assets, add "Best" variations
• Monthly: Full review, replace 20-30% of assets, seasonal updates

Replacement rules: Low >30 days → Replace. Low <30 days → Monitor. Good → Keep, test variations. Best → Keep, create similar. Pending → Wait.

Audience Signals Optimization

Audience Signals = HINTS for Google AI, NOT targeting restrictions. Google will advertise beyond your signals.

Signal priority (strongest to weakest):

1. Your Data: Customer Match (converters)
2. Your Data: Website remarketing (high intent)
3. Custom Segments: Keywords (competitor/product)
4. Custom Segments: URLs (competitor sites)
5. In-Market Audiences
6. Affinity Audiences
7. Demographics

Example configuration — Bestsellers AG:

• Your Data: All Purchasers, Product Viewers (90d), Cart Abandoners (30d)
• Custom Segments: top product names, competitor product pages
• Interests: relevant In-Market + Affinity categories

Audience signal troubleshooting (ignored signals, too broad/narrow) in references/decision-trees.md.

Search Themes

Keywords you give PMax to understand relevant search queries (hints for AI, not exact targeting).

• Recommended: 3-7 per asset group (max 25)
• Types: Branded, Generic Product, Long-tail, Problem-solving
• Update monthly: check Search Terms report, add high-performers, remove non-performers

Per-category search theme examples in references/detailed-reference.md.

Listing Groups (E-commerce)

Listing Groups determine which products belong to which Asset Group.

Segmentation attributes: Category, Brand, Condition, Item ID, Product Type, Custom Labels (0-4).

Custom Labels strategy:

• Label 0: Margin Level (High/Medium/Low)
• Label 1: Performance Tier (Bestseller/Rising Star/Long Tail)
• Label 2: Seasonal (Spring/Summer/Fall/Winter)
• Label 3: Promo Status (Full Price/On Sale/Clearance)

Optimization: Identify top 20% products (own asset group, premium signals) and bottom 20% (exclude or improve feed quality). Exclude out-of-stock, below-breakeven margin, seasonally irrelevant products.

PMax New Capabilities (2025-2026)

• Campaign-level negative keywords (v20): Add via UI or API. Protect specific queries for separate Search campaigns.
• Search terms reporting (v21): PMax search term data now in API. Use to find negatives and new keyword opportunities.
• First-party audience exclusions (Mar 2026): Exclude existing customers for pure prospecting.
• Ad network type breakdown (v22): See which network (Search, Display, YouTube) drives PMax performance.
• Demographic reporting (Mar 2026): Age/gender breakdowns for budget efficiency analysis.

Creative testing approaches (sequential, split AG, experiments) and asset refresh cadence in references/detailed-reference.md.

Output: Asset Group Optimization Template

# Asset Group Optimization Plan

## Campaign Overview
- **Campaign:** [Name] | **Account:** [Name] | **Period:** [Date range]

## Asset Group Structure
| Asset Group | Products/Focus | Budget % | ROAS | Status |
|-------------|---------------|----------|------|--------|
| [AG 1] | [Products] | [%] | [X]x | [Status] |

## Asset Performance Review
| Asset Type | Rating | Action |
|------------|--------|--------|
| [Headline/Image/Video] | [Best/Good/Low] | [Keep/Replace/Add variation] |

## Audience Signals Review
| Signal Type | Signal | Performance |
|-------------|--------|-------------|
| Your Data | [Signal] | [Performance] |
| Custom | [Signal] | [Performance] |

## Action Items
- Immediate: [This week actions]
- Short-term: [Next 2 weeks]
- Long-term: [Next month]

## Success Metrics
| Metric | Current | Target | Timeline |
|--------|---------|--------|----------|
| ROAS | [X]x | [Y]x | [Date] |
| CPA | $[X] | $[Y] | [Date] |