YouTube Optimization

Optimization Cycle:
├── Analyze
│   ├── Review key metrics
│   ├── Identify trends
│   └── Find opportunities
├── Hypothesize
│   ├── Form optimization theory
│   └── Predict impact
├── Test
│   ├── Implement changes
│   └── Control variables
├── Measure
│   ├── Wait for significance
│   └── Compare to baseline
└── Scale
    ├── Apply winners broadly
    └── Document learnings

High Impact, Quick Wins:
├── Pause low-performing ads
├── Increase bids on top audiences
├── Add negative placements
└── Refresh declining creatives

High Impact, More Effort:
├── New creative concepts
├── Audience expansion
├── Landing page optimization
└── Conversion tracking improvements

Lower Priority:
├── Minor copy changes
├── Small targeting tweaks
├── Frequency cap adjustments
└── Cosmetic reporting changes

Awareness Goals:
├── CPM Bidding
│   └── Pay per 1,000 impressions
├── Target CPM
│   └── Average CPM target
└── Maximize Impressions
    └── Most reach for budget

Consideration Goals:
├── CPV Bidding
│   └── Pay per 30s view
├── Maximum CPV
│   └── Set ceiling
└── Target CPV
    └── Average view cost

Conversion Goals:
├── Target CPA
│   └── Set cost per conversion
├── Target ROAS
│   └── Set return target
└── Maximize Conversions
    └── Most conversions for budget

Bid Adjustment Factors:
├── Device
│   ├── Mobile: +20% to -50%
│   ├── Desktop: baseline
│   └── TV: +30% (high engagement)
├── Time of Day
│   ├── Peak hours: +10%
│   └── Off-hours: -20%
├── Demographics
│   ├── Top performing age: +15%
│   └── Under-performing: -30%
└── Audiences
    ├── Remarketing: +50%
    └── Broad prospecting: baseline

Best Practices:
├── Start with automated bidding
├── Let campaigns learn (2 weeks)
├── Apply adjustments gradually
├── Monitor for bid landscape changes
└── Don't over-optimize (audience shrinkage)

Review Weekly:
├── Performance by audience
├── Overlap between audiences
├── Frequency by segment
├── Conversion rates
└── Cost per conversion

Segmentation Actions:
├── High VTR, Low CVR → Exclude
├── High CVR, Low VTR → Bid up
├── High volume, Average CVR → Test creatives
└── Low volume, High CVR → Expand similar

Expansion Strategy:
├── Phase 1: Core audiences
│   └── Remarketing, Customer Match
├── Phase 2: Similar audiences
│   └── Lookalikes of converters
├── Phase 3: In-market
│   └── Active purchase intent
├── Phase 4: Affinity
│   └── Lifestyle alignment
└── Phase 5: Broad targeting
    └── Let algorithm find audiences

Similar Audience Building:
├── Source: Email list (5K+ users)
├── Source: Purchasers (1K+ users)
├── Source: High-value segment
└── Combine with content targeting

Strategic Exclusions:
├── Existing customers (when appropriate)
├── Low-quality placements
├── Completed converters
├── Brand safety categories
└── Non-performing demographics

Placement Exclusions:
├── Review placement report weekly
├── Exclude channels with \<0.1% VTR
├── Exclude irrelevant categories
└── Build exclusion lists over time

Key Metrics:
├── View Rate: Hook effectiveness
├── Watch Time: Content quality
├── CTR: CTA effectiveness
├── CVR: Message-market fit
└── Relative Performance: vs benchmarks

Diagnostic Questions:
├── Low VTR → Is hook compelling?
├── High VTR, Low CTR → Is CTA clear?
├── High CTR, Low CVR → Landing page issue?
└── Declining metrics → Creative fatigue?

Refresh Triggers:
├── VTR drops >20% from peak
├── CPV increases >30%
├── Frequency >3 per user
├── Monthly regardless of performance
└── Seasonal relevance changes

Refresh Strategy:
├── Don't pause all ads at once
├── Rotate 1-2 new creatives weekly
├── Keep winners running
├── Archive underperformers
└── Document what works

Test Priority:
├── 1. Hook variations (biggest impact)
├── 2. Message/offer variations
├── 3. CTA variations
├── 4. Length variations
└── 5. Style/format variations

Test Structure:
├── Single variable isolation
├── Minimum 1,000 views per version
├── Run for full week minimum
├── Statistical significance required
└── Document all learnings

When to Scale:
├── Consistent performance 2+ weeks
├── CPA/ROAS within target
├── Strong creative (VTR healthy)
├── Audience headroom exists
└── Conversion tracking stable

Scaling Methods:
├── Vertical: Increase budget
│   └── 20% increments max
├── Horizontal: New audiences
│   └── Duplicate winning campaigns
└── Creative: More variations
    └── Similar to winners

Scaling Timeline:
Week 1: Baseline performance
Week 2: Confirm consistency
Week 3: +20% budget
Week 4: Evaluate impact
Week 5: +20% if stable
Week 6: Horizontal expansion

Warning Signs:
├── CPA rises >15% → Pause scaling
├── VTR drops → Check frequency
├── Volume flat at higher budget → Ceiling reached
└── New audiences underperform → Refine targeting

Low Impressions:
├── Bid too low → Increase bid
├── Audience too narrow → Expand targeting
├── Budget too low → Increase daily budget
├── Ad disapproved → Check policy
└── Competition spike → Check auction insights

Low View Rate:
├── Weak hook → Test new openers
├── Irrelevant audience → Narrow targeting
├── Ad fatigue → Refresh creative
├── Poor placement quality → Add exclusions
└── Wrong format → Test different ad types

High CPA:
├── Low intent audience → More qualified targeting
├── Weak CTA → Strengthen call-to-action
├── Landing page issues → Test new pages
├── Conversion tracking broken → Verify setup
└── Product-market fit → Review offer

Ad Disapprovals:
├── Check Google Ads Policy Center
├── Review specific disapproval reason
├── Edit and resubmit
├── Request manual review if unclear
└── Create compliant alternative

When Performance Drops:
├── Check for external factors
│   ├── Seasonality
│   ├── Competition
│   └── News/events
├── Review recent changes
│   ├── Bid adjustments
│   ├── Budget changes
│   └── Targeting changes
├── Audit tracking
│   ├── Conversion tags firing
│   ├── Website changes
│   └── Landing page status
└── Creative health check
    ├── Frequency levels
    ├── View rate trends
    └── Audience saturation