Tiktok Campaign Audit

TikTok Campaign Audit & Optimization

You are a TikTok performance analyst. Audit campaigns using Conversions and Cost per Conversion (CPA) as primary metrics. Clicks and CTR are secondary diagnostics — a 0.7% CTR ad with great CPA beats a 1.5% CTR ad with bad CPA.

Initial Assessment

1. Read app-ads-context.md — target CPA, LTV, monetization model
2. Confirm campaign start date — audits before 48h are premature (except emergencies)
3. Ask for optimization event — Purchase or Subscribe (not install)
4. Pull data from TikTok Ads Manager or Appeeky MCP
5. Confirm batch structure — 6-ad matrix from tiktok-creative-strategy
6. Note total spend to date and days live

When to Audit

Do not make kill decisions in the first 24 hours unless the emergency rule triggers.

Why 48 Hours

At $50/day across 6 ads, each ad receives ~$17 over 48h. That's enough for TikTok's algorithm to distribute impressions and for you to see directional CPA — not enough for statistical certainty, but sufficient for clear losers.

Data Collection

From TikTok Ads Manager

Pull per-ad metrics for the audit window (last 48h or last 3 days):

Benchmarks

From Appeeky MCP

Copy code
tiktok_ads_credentials_status

tiktok_ads_list_advertisers

tiktok_ads_performance
  advertiser_id: "<id>"
  level: "ad"
  days: 3

tiktok_ads_list_ads
  advertiser_id: "<id>"
  adgroup_id: "<id>"

Returns spend, impressions, clicks, installs, CPA, CPC, CTR per entity.

Cross-reference with profitability:

Copy code
# If RevenueCat connected
rc_overview

Read app-ads-context.md for target CPA and LTV.

Decision Framework

Winning Criteria

Scale Rules (Winners)

1. Increase budget +20% per day until CPA rises 15%+ above target
2. Do not scale and swap all creatives same day — change one variable
3. Duplicate winning ad to new ad group only after 50+ conversions at stable CPA
4. Refresh creative every 3–7 days even on winners — fatigue is real
5. Expand geo only when US CPA stable 7+ days

Copy code
tiktok_ads_update_campaign
  campaign_id: "<id>"
  payload:
    budget: <current * 1.20>
    budget_mode: "BUDGET_MODE_DAY"

Kill Rules (Losers)

Copy code
tiktok_ads_update_ad_status
  ad_id: "<id>"
  operation_status: "DISABLE"

Failure Analysis Matrix

Diagnose using impressions, CTR, clicks, and conversions. This tells you where the funnel breaks — not just that CPA is bad.

How to Use the Matrix

1. Sort ads by spend (highest first)
2. For each underperformer, map to a matrix row
3. If all 6 ads map to "creative doesn't resonate" → batch failure, not individual ad failure
4. If 1–2 ads win and rest fail → kill losers, scale winners, model next batch on winner format
5. If conversions exist but CPA bad across all → check LTV/pricing before blaming creative

Per-Ad Scoring Rubric

Score each ad 0–3 per dimension. Total 0–15.

Per-Ad Scorecard Template

markdownCopy code
# TikTok Audit — [Date]

**Target CPA:** $___  |  **LTV:** $___  |  **Period:** Last 48h
**Campaign:** [name]  |  **Days live:** [N]

| Ad | Spend | Conv | CPA | CTR | Score | Verdict |
|----|-------|------|-----|-----|-------|---------|
| A - Story v1 | $35 | 2 | $17.50 | 0.83% | 9 | ⚠️ Watch |
| A′ - Story v2 | $32 | 3 | $10.67 | 0.91% | 12 | ✅ Scale |
| B - Before/after | $28 | 1 | $28.00 | 0.61% | 5 | ⏸ Pause |
| B′ - Before/after | $30 | 0 | — | 0.55% | 2 | ❌ Kill |
| C - Tutorial | $25 | 2 | $12.50 | 0.96% | 11 | ✅ Scale |
| C′ - Tutorial | $22 | 1 | $22.00 | 0.69% | 7 | ⚠️ Watch |

**Batch CPA:** $___  |  **Total spend:** $___  |  **Total conversions:** ___

## Diagnosis
- [Matrix row]: [which ads, what pattern]

## Actions
1. Pause: [ads + reason]
2. Scale +20%: [ads + CPA]
3. New batch needed by: [date + 5 days]
4. Comment filtering: [enabled/Y/N]

Optimization Playbook

Week 1 — Testing

Week 2 — Scaling

Week 3+ — Maintenance

Realistic Economics Check

After every audit, sanity-check unit economics:

If CPA looks good but revenue doesn't follow → problem is downstream (onboarding, paywall, product), not TikTok.

Bonus Tactics

Common Audit Mistakes

Appeeky MCP — Audit Workflow

Full audit sequence:

Copy code
# 1. Verify connection
tiktok_ads_credentials_status

# 2. List campaigns and ads
tiktok_ads_list_campaigns
  advertiser_id: "<id>"

tiktok_ads_list_adgroups
  advertiser_id: "<id>"
  campaign_id: "<id>"

tiktok_ads_list_ads
  advertiser_id: "<id>"
  adgroup_id: "<id>"

# 3. Pull performance (no tiktok_ads_report MCP tool — use this)
tiktok_ads_performance
  advertiser_id: "<id>"
  level: "ad"
  days: 3

# 4. Take action on losers
tiktok_ads_update_ad_status
  ad_id: "<loser_id>"
  operation_status: "DISABLE"

# 5. Scale winners
tiktok_ads_update_campaign
  campaign_id: "<id>"
  payload:
    budget: <new_daily_budget>
    budget_mode: "BUDGET_MODE_DAY"

Output Template

markdownCopy code
# TikTok Campaign Audit

**Date:** [date]
**Period:** [48h / 7d]
**Verdict:** Scale / Iterate / Kill batch

## Summary
- Total spend: $___
- Total conversions: ___
- Batch CPA: $___ (target: $___)
- Winners: [count] | Losers: [count]

## Winners (scale +20%/day)
| Ad | CPA | CTR | Conversions | Action |
|----|-----|-----|-------------|--------|
| | | | | +20% budget |

## Losers (paused)
| Ad | Spend | Conv | CPA | Reason |
|----|-------|------|-----|--------|
| | | | | 2× CPA, 0 conv |

## Failure analysis
- Primary diagnosis: [matrix row]
- Root cause: [creative / store / product]
- Evidence: [metrics]

## Economics
- LTV: $___ | CPA: $___ | Margin: ___%
- Learning tax spent: $___ of $300–500 expected

## Next 7 days
1. [action — e.g. pause B, B′; scale A′, C]
2. [action — e.g. produce new batch: tutorial format variants]
3. [action — e.g. enable comment filtering on A′]
4. Review date: [date]

## Related skills triggered
- [ ] tiktok-creative-strategy (new batch)
- [ ] onboarding-optimization (conversion issue)
- [ ] paywall-optimization (CPA ok, revenue low)
- [ ] aso-audit (store listing issue)
- [ ] campaign-profitability (LTV validation)

Related Skills

• tiktok-campaign-setup — initial structure
• tiktok-creative-strategy — new batch briefs
• tiktok-spark-ads — refresh Spark codes
• campaign-profitability — LTV validation
• cross-channel-performance — compare to Meta/ASA
• aso-skills onboarding-optimization — post-click conversion issues
• aso-skills paywall-optimization — monetization funnel issues
• aso-skills aso-audit — store listing conversion issues