LeoYeAI/agent-master-skills / skills/apple-search-ads
Apple Search Ads
VerifiedCreate, optimize, and scale Apple Search Ads campaigns with API automation, attribution integration, and bid strategy recommendations.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 18.7 KB |
| _meta.json | config | 289 B |
| api-reference.md | markdown | 7.8 KB |
| ios-integration.md | markdown | 11.0 KB |
| memory-template.md | markdown | 3.1 KB |
| scripts.md | markdown | 16.0 KB |
| setup.md | markdown | 1.5 KB |
| strategy.md | markdown | 8.7 KB |
strategy.md
View on GitHub# Strategy Playbook — Apple Search Ads
Complete strategic guide for planning, launching, and scaling ASA campaigns.
## Campaign Planning
### Before You Start
**Prerequisites checklist:**
- [ ] App Store listing optimized (screenshots, description, keywords)
- [ ] Conversion tracking implemented (AdServices or MMP)
- [ ] LTV model estimated (even rough)
- [ ] Budget allocated ($500+ monthly recommended to start)
- [ ] Competitor research done
### Know Your Numbers
| Metric | How to Calculate | Why It Matters |
|--------|------------------|----------------|
| **LTV** | Revenue per user over lifetime | Sets max CPA |
| **Target CPA** | LTV / 3 to LTV / 5 | Profitable acquisition |
| **Breakeven CPA** | LTV - margins | Maximum spend |
| **TTR benchmark** | 5-10% | Ad relevance |
| **CVR benchmark** | 30-60% | Landing page quality |
## Campaign Architecture
### The 4-Campaign Structure
```
1. BRAND (Defend)
└── Your app name, brand terms
└── Goal: 100% impression share, lowest CPA
└── Budget: 10-20% of total
2. CATEGORY (Capture)
└── Generic category terms
└── Goal: Volume with target CPA
└── Budget: 40-50% of total
3. COMPETITOR (Steal)
└── Competitor app names
└── Goal: Profitable poaching
└── Budget: 15-25% of total
4. DISCOVERY (Learn)
└── Search Match enabled
└── Goal: Find new keywords
└── Budget: 10-15% of total
```
### Ad Group Structure
For each campaign, create ad groups by:
**Option A: By Match Type**
```
Campaign: MyApp - US - Category
├── Ad Group: Category - Exact
│ └── meditation app, mindfulness, daily meditation
├── Ad Group: Category - Broad
│ └── meditation, mindfulness
└── Ad Group: Category - Discovery
└── Search Match ON
```
**Option B: By Theme**
```
Campaign: MyApp - US - Category
├── Ad Group: Meditation
│ └── meditation app, best meditation, guided meditation
├── Ad Group: Sleep
│ └── sleep app, sleep sounds, better sleep
└── Ad Group: Stress
└── stress relief, anxiety app, calm app
```
## Keyword Strategy
### Keyword Research Process
1. **Start with App Store Connect**
- Download search terms from your organic sources
- These are proven to have intent for your app
2. **Competitor Analysis**
- What keywords do competitors rank for?
- Use tools: Sensor Tower, App Annie, Mobile Action
3. **Apple Search Ads Suggestions**
- Use the keyword suggestions in ASA dashboard
- Add with broad match first to test
4. **Search Term Mining**
- Review Search Term reports weekly
- Graduate high performers to exact match
### Keyword Tiers
| Tier | Description | Bid Strategy |
|------|-------------|--------------|
| **Tier 1** | Brand terms | Bid high (2-3x avg) |
| **Tier 2** | Category leaders | Bid at target CPA |
| **Tier 3** | Long-tail specific | Bid below avg, test |
| **Tier 4** | Competitor names | Start low, test carefully |
| **Tier 5** | Discovery/Broad | Lowest bids, mine data |
### Negative Keywords
**Add negatives immediately for:**
- "free" (if you're paid)
- Competitor names (in brand campaign)
- Unrelated categories
- Misspellings that waste budget
**Negative keyword structure:**
```
Campaign-level negatives:
└── Broad terms you NEVER want
Ad Group-level negatives:
└── Terms that belong in OTHER ad groups
```
## Bidding Strategy
### Starting Bids
| Campaign Type | Starting Bid | Adjust After |
|---------------|--------------|--------------|
| Brand | $2-3 | 7 days |
| Category | $1-2 | 7 days |
| Competitor | $0.50-1 | 14 days |
| Discovery | $0.30-0.50 | 7 days |
### Optimization Cadence
**Daily (5 min):**
- Check spend vs budget
- Pause anything wildly over CPA
**Weekly (30 min):**
- Review search terms
- Add negatives
- Graduate keywords
- Adjust bids ±20%
**Monthly (2 hours):**
- Full performance review
- Budget reallocation
- Test new keywords
- Analyze competitor trends
### Bid Adjustment Rules
```
If CPA < Target by 20%+ AND volume is good:
→ Increase bid 20%
If CPA > Target by 20%+:
→ Decrease bid 20%
→ Or pause if no improvement after 2 weeks
If Impressions low BUT CPA is good:
→ Increase bid gradually to get more volume
If TTR < 3%:
→ Ad isn't relevant, lower bid or pause
```
## Scaling Strategy
### Phase 1: Validation ($500-2K/month)
Goals:
- Prove attribution works
- Find CPA benchmarks
- Identify winning keywords
Structure:
- 1-2 campaigns (Brand + Category)
- 3-5 ad groups
- 20-50 keywords
- US only
### Phase 2: Optimization ($2K-10K/month)
Goals:
- Optimize to target CPA
- Build keyword portfolio
- Test competitors
Structure:
- 4 campaigns (full structure)
- 10-15 ad groups
- 100-200 keywords
- Add UK, Canada, Australia
### Phase 3: Scale ($10K+/month)
Goals:
- Maximize volume at CPA
- International expansion
- Custom Product Pages
Structure:
- Multiple campaigns per country
- Dedicated teams for optimization
- Advanced automation
- All major markets
### Budget Scaling Rules
**Don't scale too fast.** Increase budget by max 20-30% at a time.
```
Week 1: $50/day
Week 2: If CPA good → $65/day
Week 3: If CPA good → $85/day
Week 4: If CPA good → $110/day
...
```
If CPA spikes after increase, pull back immediately.
## Multi-Country Expansion
### Expansion Order (English-first)
1. **Tier 1:** US (highest volume, test here first)
2. **Tier 2:** UK, Canada, Australia (English, easy)
3. **Tier 3:** Germany, France (high value, need localization)
4. **Tier 4:** Japan, South Korea (high LTV, need localization)
5. **Tier 5:** Brazil, Mexico, India (volume play, lower CPT)
### Localization Checklist
For each new country:
- [ ] App Store listing translated
- [ ] Screenshots localized
- [ ] Keywords researched in local language
- [ ] Custom Product Page created
- [ ] Separate campaign created
- [ ] Local currency budget set
### Country-Specific Considerations
| Country | Notes |
|---------|-------|
| US | Highest competition, highest CPT, but most volume |
| UK | Similar to US but smaller, good for testing |
| Germany | High LTV, privacy-conscious, need localization |
| Japan | Very high LTV for games, unique keywords |
| Brazil | High volume, low CPT, Portuguese required |
## Custom Product Pages
### Strategy
Create CPPs for:
1. **Different audiences** — Feature focus (workout vs meditation)
2. **Different intents** — Problem vs aspiration
3. **Seasonal** — Holidays, New Year, summer
### A/B Testing
```
Campaign: MyApp - US - Category
├── Ad Group: Meditation - Control
│ └── Default Product Page
│ └── 50% traffic
└── Ad Group: Meditation - Test
└── CPP: Sleep-focused
└── 50% traffic
```
Run for 2 weeks minimum, then compare CVR.
## Advanced Tactics
### Dayparting
Test performance by hour of day:
1. Run 2 weeks with all hours
2. Analyze by hour in reports
3. Turn off worst-performing hours
4. Reallocate budget to best hours
### Audience Exclusions
Exclude users who already have your app:
```json
"targetingDimensions": {
"appDownloaders": {
"excluded": [YOUR_ADAM_ID]
}
}
```
### Search Tab Campaigns
Lower intent but cheaper clicks:
- Good for brand awareness
- Use different creative messaging
- Expect lower CVR but lower CPT
- Separate budget from Search Results
### Today Tab Campaigns
Premium placement, brand awareness:
- Higher minimum budget
- Best for app launches
- Seasonal promotions
- Brand building
## Reporting & Analysis
### Key Reports to Run
| Report | Frequency | Purpose |
|--------|-----------|---------|
| Campaign performance | Daily | Monitor spend/CPA |
| Search terms | Weekly | Keyword mining |
| Ad group performance | Weekly | Bid optimization |
| Impression share | Monthly | Competitive view |
| Geo performance | Monthly | Country decisions |
### Funnel Analysis
Track the full funnel:
```
Impressions → Taps → Installs → Activations → Revenue
↓ ↓ ↓ ↓ ↓
TTR CVR Retention Conversion ROAS
```
### Attribution Windows
Default: 30-day click-through attribution
Consider:
- Short window (7 days) for quick decisions
- Long window (30 days) for accurate LTV
## Common Mistakes
1. **Starting with broad match only** → No control, wasted spend
2. **Ignoring Search Terms** → Missing optimization gold
3. **Same bid everywhere** → Different keywords have different values
4. **Scaling too fast** → CPA spikes, budget burned
5. **No negatives** → Paying for irrelevant searches
6. **Mixing countries** → Can't optimize properly
7. **Not defending brand** → Competitors steal your users
8. **Ignoring CVR** → Problem might be App Store, not ads
9. **No attribution** → Optimizing blind
10. **Copying competitors** → Their strategy isn't yours
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirements | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |