{"id":13316,"plugin_id":"plugin_asdk_app_698a0208e7d881919cc8d52147cb0331","kind":"skill","collection_source":null,"comparison_source":null,"observed_at":"2026-09-30T23:07:18.508Z","digest":"f4027c12fdefc0b8884d1e43e9309ab1739a9abb756e6951370f555dca867698","against":null,"payload":{"description":"This skill should be used when the user asks to \"choose a Google Ads bid strategy\", \"compare tCPA vs tROAS\", \"set up value-based bidding\", \"migrate from manual to Smart Bidding\", or mentions \"Portfolio Bidding\", \"Maximize Conversions\", or \"learning phase management\". Do NOT use for: Meta Ads bidding (use meta-bid-strategy-selector), LinkedIn bidding (use linkedin-bid-strategy-selector), keyword strategy (use keyword-strategy-planner).","included_files":[],"name":"google-ads-bid-strategy-selector","skill_md_contents":"---\nname: google-ads-bid-strategy-selector\ndescription: \"This skill should be used when the user asks to \\\"choose a Google Ads bid strategy\\\", \\\"compare tCPA vs tROAS\\\", \\\"set up value-based bidding\\\", \\\"migrate from manual to Smart Bidding\\\", or mentions \\\"Portfolio Bidding\\\", \\\"Maximize Conversions\\\", or \\\"learning phase management\\\". Do NOT use for: Meta Ads bidding (use meta-bid-strategy-selector), LinkedIn bidding (use linkedin-bid-strategy-selector), keyword strategy (use keyword-strategy-planner).\"\n---\n# Bid Strategy Selector\n\nComplete guide for choosing and implementing the right Google Ads Smart Bidding strategy based on goals, data, and account situation.\n\n## Quick Decision Tree\n\n```\nWHICH BID STRATEGY IS RIGHT FOR YOU?\n│\n├── NEW ACCOUNT / LOW DATA (<30 conversions/month)\n│   └── MAXIMIZE CONVERSIONS (no target)\n│       └── Goal: Collect data, complete learning phase\n│\n├── LEAD GENERATION with known lead value\n│   ├── Consistent conversion volume (50+/month)?\n│   │   └── YES → TARGET CPA\n│   │   └── NO → MAXIMIZE CONVERSIONS\n│   └── Variable lead values?\n│       └── YES → MAXIMIZE CONVERSION VALUE + tROAS\n│\n├── E-COMMERCE with purchase tracking\n│   ├── Focus on volume (market share)?\n│   │   └── MAXIMIZE CONVERSION VALUE\n│   ├── Focus on profitability?\n│   │   └── TARGET ROAS\n│   └── Balance both?\n│       └── MAXIMIZE CONVERSION VALUE + tROAS target\n│\n└── MULTIPLE CAMPAIGNS with same goal\n    └── PORTFOLIO BID STRATEGY\n        └── Shared strategy across campaigns\n```\n\n## Smart Bidding Overview\n\n```\nSMART BIDDING COMPARISON\n════════════════════════\n\n┌─────────────────────────┬───────────┬───────────┬─────────────────────┐\n│ STRATEGY                │ CONTROL   │ DATA REQ  │ BEST FOR            │\n├─────────────────────────┼───────────┼───────────┼─────────────────────┤\n│ Maximize Conversions    │ None      │ Low       │ New accounts,       │\n│ (without target)        │           │           │ data collection     │\n├─────────────────────────┼───────────┼───────────┼─────────────────────┤\n│ Maximize Conversions    │ CPA cap   │ Medium    │ Lead gen with       │\n│ + Target CPA            │           │ (50+/mo)  │ cost constraints    │\n├─────────────────────────┼───────────┼───────────┼─────────────────────┤\n│ Maximize Conv. Value    │ None      │ Low       │ E-commerce volume,  │\n│ (without target)        │           │           │ initial learning    │\n├─────────────────────────┼───────────┼───────────┼─────────────────────┤\n│ Maximize Conv. Value    │ ROAS      │ Medium    │ E-commerce profit,  │\n│ + Target ROAS           │ target    │ (50+/mo)  │ scaling             │\n├─────────────────────────┼───────────┼───────────┼─────────────────────┤\n│ Manual CPC              │ Max CPC   │ None      │ Niche, B2B,         │\n│ (Enhanced CPC opt.)     │ per kw    │           │ small accounts      │\n└─────────────────────────┴───────────┴───────────┴─────────────────────┘\n```\n\n## Maximize Conversions\n\n### How It Works\n\n```\nMAXIMIZE CONVERSIONS ENGINE\n===========================\n\n┌────────────────────────────────────────────────────────────────┐\n│  GOOGLE AI OPTIMIZES FOR:                                      │\n│  Maximum number of conversions within your daily budget        │\n│                                                                │\n│  SIGNALS USED:                                                 │\n│  ├── Device, location, time of day                             │\n│  ├── Browser, OS, demographics                                 │\n│  ├── Search query and intent signals                           │\n│  ├── Remarketing lists membership                              │\n│  ├── Historical conversion patterns                            │\n│  └── Real-time auction dynamics                                │\n│                                                                │\n│  YOU CONTROL:                                                  │\n│  ├── Daily budget (spending limit)                             │\n│  ├── Target CPA (optional, as constraint)                      │\n│  └── Conversion actions (which ones to optimize for)           │\n└────────────────────────────────────────────────────────────────┘\n```\n\n### When to Use Maximize Conversions\n\n```\nUSE MAXIMIZE CONVERSIONS WHEN:\n──────────────────────────────\n• New account with little historical data\n• First 2-4 weeks of a new campaign\n• Lead generation focus (single conversion type)\n• Budget is more important than CPA efficiency\n• Collecting data for later tCPA transition\n\nDO NOT USE WHEN:\n────────────────\n• Strict CPA requirements (use tCPA)\n• E-commerce with purchase values (use Max Conv Value)\n• Very low budget (<EUR20/day) - too few learnings\n• Campaign with multiple conversion types without a primary one\n```\n\n### Maximize Conversions + Target CPA\n\n```\nADDING TARGET CPA\n=================\n\nWHEN:\n├── 50+ conversions in the past 30 days\n├── Stable performance (no major fluctuations)\n├── Known target CPA (break-even or goal)\n└── After successful pure Maximize Conversions phase\n\nCALCULATING TARGET CPA:\n───────────────────────\nBreak-even CPA (Lead Gen):\n└── Lead Value x Conversion Rate to Sale\n\nExample:\n├── Lead value (as sale): EUR500\n├── Close rate: 10%\n├── Break-even CPA: EUR500 x 0.10 = EUR50\n\nStarting Target CPA:\n├── Week 1-2: 120% of break-even (EUR60)\n├── Week 3-4: 110% of break-even (EUR55)\n├── Week 5+: 100% or tighter if stable\n\nWARNING: NEVER start BELOW your historical average CPA!\n```\n\n## Maximize Conversion Value\n\n### How It Works\n\n```\nMAXIMIZE CONVERSION VALUE ENGINE\n================================\n\n┌────────────────────────────────────────────────────────────────┐\n│  GOOGLE AI OPTIMIZES FOR:                                      │\n│  Maximum total conversion value within your daily budget       │\n│                                                                │\n│  REQUIREMENTS:                                                 │\n│  ├── Conversion tracking with VALUE (purchase value)           │\n│  ├── Accurate revenue/value data                               │\n│  └── Consistent value tracking                                 │\n│                                                                │\n│  AI PRIORITIZES:                                               │\n│  ├── High-value transactions over low-value ones               │\n│  ├── Users with high predicted value                           │\n│  └── Queries that historically generate high values            │\n└────────────────────────────────────────────────────────────────┘\n```\n\n### Maximize Conversion Value + Target ROAS\n\n```\nADDING TARGET ROAS\n==================\n\nWHEN:\n├── 50+ conversions with value in the past 30 days\n├── Consistent value tracking (no gaps)\n├── Known break-even or target ROAS\n└── After successful pure Max Conv Value phase\n\nCALCULATING TARGET ROAS:\n────────────────────────\nBreak-even ROAS = 1 / Profit Margin\n\nExample:\n├── Profit margin: 40%\n├── Break-even ROAS: 1 / 0.40 = 2.5 (250%)\n├── Spending EUR100 means generating EUR250 in revenue is needed\n\nStarting Target ROAS:\n├── Week 1-2: 80% of break-even (200% if break-even is 250%)\n├── Week 3-4: 90% of break-even (225%)\n├── Week 5+: 100% or tighter if stable\n\nWARNING: Too aggressive a ROAS target = no delivery!\n```\n\n## Portfolio Bid Strategies\n\n### What Are Portfolio Strategies?\n\n```\nPORTFOLIO BID STRATEGY\n======================\n\n= One bid strategy shared across multiple campaigns\n\nADVANTAGES:\n├── More data for learning → better optimization\n├── Centralized bid management\n├── Budget flexibility across campaigns\n└── Better performance for small campaigns\n\nLIMITATIONS:\n├── All campaigns must share the same goal\n├── Shared learning can be suboptimal per campaign\n└── Less granular control\n```\n\n### Portfolio Strategy Setup\n\n```\nPORTFOLIO STRATEGY TYPES\n========================\n\n1. TARGET CPA PORTFOLIO\n   └── Multiple Search/Display campaigns, same CPA goal\n   └── Example: Brand + Non-brand Search\n\n2. TARGET ROAS PORTFOLIO\n   └── E-commerce campaigns with same margin target\n   └── Example: Shopping + Search + PMax\n\n3. MAXIMIZE CONVERSIONS PORTFOLIO\n   └── Aggregate data for faster learning\n   └── Example: Bundle new campaigns\n\n4. TARGET IMPRESSION SHARE PORTFOLIO\n   └── Brand visibility campaigns\n   └── Example: Branded Search campaigns\n\nSETUP LOCATION:\nTools & Settings → Shared Library → Bid Strategies\n```\n\n### When to Use Portfolio\n\n```\nPORTFOLIO DECISION MATRIX\n=========================\n\nUSE PORTFOLIO WHEN:\n├── Multiple campaigns with <50 conversions/month each\n├── Campaigns have exactly the same KPI targets\n├── A single point for bid management is desired\n└── Small budgets spread across multiple campaigns\n\nUSE INDIVIDUAL WHEN:\n├── Campaigns have different margin/CPA targets\n├── Sufficient conversions per campaign (50+/month)\n├── Different product types/audiences\n└── Need for campaign-level bid adjustments\n```\n\n## Learning Phase Management\n\n### Learning Phase Basics\n\n```\nLEARNING PHASE EXPLAINED\n=========================\n\nWHAT:\n├── Period during which Smart Bidding collects data\n├── Bids can fluctuate\n├── Performance may temporarily worsen\n└── DO NOT intervene during this phase\n\nDURATION:\n├── Typical: 7-14 days\n├── Requirement: ~50 conversions (or actions)\n├── Can take longer with low volume\n└── Status visible in campaign UI\n\nLEARNING PHASE STATUS:\n├── \"Learning\" = Actively learning\n├── \"Learning (limited)\" = Insufficient data\n├── \"Eligible\" = Learning complete\n└── \"Limited\" = Other issue (budget, etc.)\n```\n\n### What Resets the Learning Phase?\n\n```\nACTIONS THAT RESET LEARNING\n============================\n\nAVOID THESE DURING LEARNING:\n─────────────────────────────\n• Changing bid strategy\n• Adjusting Target CPA/ROAS (>20%)\n• Changing conversion action\n• Increasing or decreasing budget >20%\n• Pausing campaign for >7 days\n\nSAFE DURING LEARNING:\n─────────────────────\n• Adding or pausing ads\n• Adding keywords (small batches)\n• Adding negatives\n• Budget changes <20%\n• Ad copy adjustments\n```\n\n### Learning Phase Troubleshooting\n\n```\nLEARNING PHASE ISSUES\n=====================\n\nPROBLEM: \"Learning (limited)\" stays stuck\n──────────────────────────────────────────\nCause: Insufficient conversions\nSolutions:\n├── Increase budget\n├── Broader targeting (more volume)\n├── Higher-funnel conversion action (temporarily)\n└── Wait longer (sometimes needed)\n\nPROBLEM: CPA spikes during learning\n─────────────────────────────────────\nThis is normal! The AI is testing boundaries.\nActions:\n├── DO NOT panic\n├── Wait at least 7-10 days\n├── Monitor the trend, not daily CPA\n└── If >14 days poor: evaluate targeting/budget\n\nPROBLEM: Learning takes >3 weeks\n─────────────────────────────────\nPossible causes:\n├── Insufficient budget\n├── Too niche targeting\n├── Poor ad quality\n└── Tracking issues\n```\n\n## Value-Based Bidding\n\n### Value Rules (2025+)\n\n```\nVALUE RULES EXPLAINED\n=====================\n\nWHAT:\n├── Dynamically adjust conversion values\n├── Based on user/context signals\n├── AI bids higher for high-value segments\n└── Available for all Smart Bidding strategies\n\nAVAILABLE SIGNALS:\n├── Device (mobile, desktop, tablet)\n├── Location (geographic)\n├── Audience (Customer Match, remarketing)\n└── Time (planned for future)\n\nEXAMPLE SETUP:\n──────────────\nValue Rule 1: High-Value Customers\n├── Condition: Customer Match list = \"VIP Customers\"\n├── Adjustment: +50% value\n└── Effect: EUR100 purchase → EUR150 for bidding\n\nValue Rule 2: Low-Intent Location\n├── Condition: Location = \"Low converting region\"\n├── Adjustment: -30% value\n└── Effect: EUR100 purchase → EUR70 for bidding\n```\n\n### New Customer Acquisition\n\n```\nNEW CUSTOMER BIDDING (2025+)\n============================\n\nLOCATION: Campaign Settings → Customer Acquisition\n\nOPTIONS:\n├── Bid higher for new customers: +X% bid adjustment\n├── Only bid for new customers: Exclude existing\n└── No differentiation (default)\n\nSETUP REQUIREMENTS:\n├── Customer Match list of existing customers\n├── Conversion tracking active\n└── Sufficient new vs returning data\n\nRECOMMENDED START:\n├── +20% for new customers\n├── Monitor new vs returning ROAS\n├── Adjust based on LTV data\n└── E-commerce: Consider first-purchase margin\n```\n\n## Bid Strategy Migration\n\n### From Manual to Smart Bidding\n\n```\nMANUAL → SMART BIDDING MIGRATION\n================================\n\nSTEP 1: PREPARATION (Week -2 to -1)\n────────────────────────────────────\n□ Verify conversion tracking\n□ Minimum 30 conversions/month\n□ Document baseline metrics\n□ Set budget (min EUR50/day)\n\nSTEP 2: INITIAL SETUP (Week 1)\n──────────────────────────────\n□ Start with Maximize Conversions (no target)\n□ Expect fluctuations\n□ DO NOT intervene\n\nSTEP 3: MONITORING (Week 2-3)\n─────────────────────────────\n□ Monitor learning phase\n□ Compare with baseline\n□ Still DO NOT intervene\n\nSTEP 4: OPTIMIZATION (Week 4+)\n──────────────────────────────\n□ Evaluate performance vs manual\n□ Add target if stable\n□ Start conservative (120% of achieved)\n```\n\n### Strategy Switch Checklist\n\n```\nBID STRATEGY SWITCH PROTOCOL\n============================\n\n□ PRE-SWITCH:\n├── Document current performance (7-day average)\n├── Calculate target (CPA/ROAS)\n├── Choose switching moment (not during peak)\n└── Prepare stakeholders (temporary fluctuations)\n\n□ DURING SWITCH:\n├── Implement new strategy\n├── Conservative target (120% of current)\n├── Screenshot for reference\n└── Set calendar reminder for review\n\n□ POST-SWITCH (Week 1-2):\n├── Daily monitoring (but no changes)\n├── Check learning phase status\n├── Compare week-over-week (not day-over-day)\n└── Note anomalies\n\n□ POST-SWITCH (Week 3+):\n├── Formal performance review\n├── Tighten targets if stable (+10%)\n├── Document learnings\n└── Continue monitoring\n```\n\n## Campaign Type Specific Recommendations\n\n### Search Campaigns\n\n```\nSEARCH BID STRATEGY GUIDE\n=========================\n\nBRAND SEARCH:\n├── Strategy: Maximize Conversions or Manual CPC\n├── Reason: High CTR, low competition\n├── Target: Impression Share >90%\n└── Note: Don't overbid — brand terms win anyway\n\nNON-BRAND SEARCH:\n├── New: Maximize Conversions (2-3 weeks)\n├── Then: Target CPA/ROAS\n├── Reason: Competitive, need efficiency\n├── Note: Broad match + Smart Bidding = Google's recommended combo\n└── AI Max: Enable AI Max for Search to unlock Search Term Matching,\n    URL Expansion, and Text Customization within existing Search campaigns\n    (Campaign.ai_max_setting.enable_ai_max)\n\nDSA (Dynamic Search Ads):\n├── Strategy: Maximize Conversions\n├── Reason: Discovery, volume focus\n├── Transition to tCPA once winning queries are known\n└── Note: Negative keywords management\n```\n\n### Shopping & PMax\n\n```\nSHOPPING / PMAX BID STRATEGY\n============================\n\nSTANDARD SHOPPING (if still used):\n├── Start: Maximize Clicks (data collection)\n├── Transition: Target ROAS after 50+ purchases\n├── Note: Product-level bidding via priorities\n\nPERFORMANCE MAX:\n├── E-commerce: Maximize Conversion Value + tROAS\n├── Lead Gen: Maximize Conversions + tCPA\n├── New: Without target (2-3 weeks)\n└── Note: PMax needs more data than Search\n\nROAS TARGETS FOR PMAX:\n├── Conservative start: 200-300%\n├── Moderate: 300-500%\n├── Aggressive: 500%+\n└── Adjust based on margin and goals\n```\n\n### Display & Video\n\n```\nDISPLAY / VIDEO BID STRATEGY\n============================\n\nDISPLAY CAMPAIGNS:\n├── Remarketing: Maximize Conversions + tCPA\n├── Prospecting: Maximize Conversions (volume focus)\n├── Brand: Target CPM (if available)\n└── Note: Longer learning due to lower volume\n\nVIDEO CAMPAIGNS:\n├── Awareness: Target CPM or Maximize Impressions\n├── Consideration: Target CPV (Cost-per-View)\n├── Conversion: Maximize Conversions\n└── Note: Video ads convert indirectly\n\nDEMAND GEN (replaced Discovery in 2025):\n├── Strategy: Maximize Conversions or tCPA\n├── Target CPC: Also available for Demand Gen (v22 addition)\n├── Reason: Hybrid awareness/conversion\n└── Note: Factor in view-through conversions\n```\n\n## Smart Bidding Exploration (v21+)\n\n```\nSMART BIDDING EXPLORATION\n=========================\n\nWHAT:\n├── Google's feature to test bid variations beyond your tROAS target\n├── API field: target_roas_tolerance_percent_millis\n├── Lets the algorithm explore auctions outside the strict target\n└── Goal: Find incremental volume while staying near your target\n\nWHEN TO ENABLE:\n├── Campaign hitting tROAS target but with limited volume\n├── Goal is to test scale without fully loosening the target\n├── Available for tROAS campaigns with sufficient conversion data (30+/mo)\n\nHOW TO CHECK:\n──────────────\ngoogle_ads_run_gaql(query=\"\n  SELECT\n    campaign.name,\n    campaign.maximize_conversion_value.target_roas,\n    campaign.maximize_conversion_value.target_roas_tolerance_percent_millis\n  FROM campaign\n  WHERE campaign.advertising_channel_type = 'SEARCH'\n    AND campaign.status = 'ENABLED'\n    AND segments.date DURING LAST_30_DAYS\n\")\n```\n\n## Performance Monitoring Script\n\n```javascript\n/**\n * Bid Strategy Performance Monitor\n *\n * Monitors Smart Bidding performance and learning phase status.\n *\n * Setup:\n * 1. Update EMAIL\n * 2. Schedule daily at 9:00\n */\n\nvar CONFIG = {\n  EMAIL: 'you@example.com',\n  CPA_THRESHOLD: 0.25,     // Alert on 25% CPA increase\n  ROAS_THRESHOLD: 0.20,    // Alert on 20% ROAS decline\n  LEARNING_DAYS_ALERT: 14  // Alert if learning >14 days\n};\n\nfunction main() {\n  var campaigns = AdsApp.campaigns()\n    .withCondition('Status = ENABLED')\n    .get();\n\n  var alerts = [];\n  var learningCampaigns = [];\n\n  while (campaigns.hasNext()) {\n    var campaign = campaigns.next();\n    var bidStrategy = campaign.getBiddingStrategyType();\n\n    // Check learning phase (via status indicators)\n    var status = checkCampaignStatus(campaign);\n    if (status.isLearning) {\n      learningCampaigns.push({\n        name: campaign.getName(),\n        strategy: bidStrategy,\n        days: status.learningDays\n      });\n    }\n\n    // Check performance changes\n    var perfAlerts = checkPerformance(campaign);\n    alerts = alerts.concat(perfAlerts);\n  }\n\n  // Send summary\n  if (alerts.length > 0 || learningCampaigns.length > 0) {\n    sendSummaryEmail(alerts, learningCampaigns);\n  }\n\n  Logger.log('Monitor complete. Alerts: ' + alerts.length);\n  Logger.log('Campaigns in learning: ' + learningCampaigns.length);\n}\n\nfunction checkCampaignStatus(campaign) {\n  // Note: Learning phase status not directly available via API\n  // This is a proxy check\n  var stats7d = campaign.getStatsFor('LAST_7_DAYS');\n  var stats14d = campaign.getStatsFor('LAST_14_DAYS');\n\n  var conv7d = stats7d.getConversions();\n  var conv14d = stats14d.getConversions();\n\n  // If <50 conversions in 14 days, likely still learning\n  return {\n    isLearning: conv14d < 50,\n    learningDays: conv14d < 50 ? 14 : 0\n  };\n}\n\nfunction checkPerformance(campaign) {\n  var alerts = [];\n  var name = campaign.getName();\n\n  var currentStats = campaign.getStatsFor('LAST_7_DAYS');\n  var previousStats = campaign.getStatsFor('LAST_14_DAYS');\n\n  var currentCPA = currentStats.getConversions() > 0 ?\n    currentStats.getCost() / currentStats.getConversions() : 0;\n\n  // Calculate previous period CPA\n  var prevConv = previousStats.getConversions() - currentStats.getConversions();\n  var prevCost = previousStats.getCost() - currentStats.getCost();\n  var previousCPA = prevConv > 0 ? prevCost / prevConv : 0;\n\n  if (previousCPA > 0 && currentCPA > 0) {\n    var change = (currentCPA - previousCPA) / previousCPA;\n    if (change > CONFIG.CPA_THRESHOLD) {\n      alerts.push({\n        campaign: name,\n        metric: 'CPA',\n        previous: previousCPA.toFixed(2),\n        current: currentCPA.toFixed(2),\n        change: (change * 100).toFixed(1) + '%'\n      });\n    }\n  }\n\n  return alerts;\n}\n\nfunction sendSummaryEmail(alerts, learningCampaigns) {\n  var subject = 'Smart Bidding Status - ' + AdsApp.currentAccount().getName();\n  var body = 'Smart Bidding Daily Report\\n';\n  body += '===========================\\n\\n';\n\n  if (learningCampaigns.length > 0) {\n    body += 'CAMPAIGNS IN LEARNING:\\n';\n    for (var i = 0; i < learningCampaigns.length; i++) {\n      var lc = learningCampaigns[i];\n      body += '- ' + lc.name + ' (' + lc.strategy + ')\\n';\n    }\n    body += '\\n';\n  }\n\n  if (alerts.length > 0) {\n    body += 'PERFORMANCE ALERTS:\\n';\n    for (var j = 0; j < alerts.length; j++) {\n      var alert = alerts[j];\n      body += '- ' + alert.campaign + ': ' + alert.metric + ' changed ';\n      body += alert.previous + ' -> ' + alert.current + ' (' + alert.change + ')\\n';\n    }\n  }\n\n  MailApp.sendEmail(CONFIG.EMAIL, subject, body);\n}\n```\n\n## Output: Bid Strategy Recommendation Template\n\n```markdown\n# Bid Strategy Recommendation\n\n## Account Situation\n- **Account type:** [E-commerce / Lead Gen / Hybrid]\n- **Monthly budget:** EUR[X]\n- **Current conversions/month:** [X]\n- **Current CPA/ROAS:** EUR[X] / [X]%\n- **Primary goal:** [Volume / Efficiency / Profitability]\n\n## Recommended Strategy\n**[STRATEGY NAME]**\n\n### Why This Strategy\n1. [Reason 1 - based on account situation]\n2. [Reason 2 - based on goals]\n3. [Reason 3 - based on data availability]\n\n### Implementation Plan\n\n**Week 1-2: Setup & Learning**\n- Switch to [strategy]\n- Target: [None / EURX / X%] (conservative)\n- Budget: EUR[X]/day\n- Action: Monitor only, no changes\n\n**Week 3-4: Evaluation**\n- Learning phase check\n- Performance vs baseline\n- Target adjustment: [Specify]\n\n**Week 5+: Optimization**\n- Tighten target to [X]\n- Continue monitoring\n- Evaluate scale opportunities\n\n### Targets\n- Primary: [CPA EURX / ROAS X%]\n- Secondary: [Conversions, Value, etc.]\n\n### Expected Results\n- CPA change: [+/- X%]\n- Volume change: [+/- X%]\n- Learning phase duration: [X weeks]\n\n### Risks & Mitigation\n- Risk: [Describe]\n- Mitigation: [Plan]\n```\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}