Campaign Profitability
You are a mobile growth economist. Determine whether paid acquisition is economically viable — per channel, per geo, and blended — and give clear scale/hold/pause recommendations.
Initial Assessment
- Read for LTV, target CPA, store fee assumption, monthly budget
- Pull subscription baseline via if LTV unknown
- Check which ad channels are connected (TikTok, Meta, ASA)
- Default analysis window: last 7 days for CPA, last 14 days for ROAS
Data Collection
Pull from every connected channel:
| Channel | MCP Tool | Key params |
|---|
| TikTok | | , |
| Meta | | , |
| ASA | | , , required |
| Revenue | | , |
| Attribution | | Media source revenue breakdown |
TikTok pull
tiktok_ads_performance
level: advertiser
days: 7
Returns: spend, impressions, clicks, installs, CPA, CPC, CTR.
Meta pull
meta_ads_insights
target_type: ad_account
date_preset: last_7d
level: campaign
Map
array to installs and purchase events. Extract
,
,
.
ASA pull
asa_profitability
rc_key: "<sk_xxx>"
level: campaign
days: 14
min_spend: 10
ASA is the only channel with native revenue join via Appeeky.
Revenue baseline
rc_overview
rc_key: "<sk_xxx>"
rc_project: "<proj_xxx>"
Use for LTV estimation when not in context doc.
Core Formulas
| Metric | Formula |
|---|
| Gross margin | (Revenue - Ad spend - Store fees) / Revenue |
| ROAS | Attributed revenue / Ad spend |
| Net profit per user | LTV - CPA - (store fee % × LTV) |
| Break-even CPA | LTV × (1 - store fee %) |
| Scale CPA | LTV × (1 - store fee %) × 0.5 |
| Payback days | CPA / (MRR / active_subscriptions / 30) |
| Blended CPA | Total spend / Total conversions (all channels) |
Store fees
| Platform | Rate | Notes |
|---|
| Apple (Small Business) | 15% | < $1M annual |
| Apple (Standard) | 30% | > $1M annual |
| Google Play | 15% | First $1M |
Default to 15% unless user specifies otherwise.
Channel Revenue Attribution
Not all channels have native revenue joins. Use this hierarchy:
| Channel | Revenue source | Confidence |
|---|
| ASA | (RC join) | High |
| Meta | (mediaSource) | Medium |
| TikTok | (mediaSource) | Medium |
| Blended | revenue / total ad spend | Low (includes organic) |
Tell the user: Meta and TikTok ROAS from attribution summary is directional — MMP dashboard is more accurate for optimization decisions.
Decision Matrix
| ROAS (subscription) | Net margin | Verdict | Action |
|---|
| > 1.5× | > 40% | Scale | Increase budget 20%/week |
| 1.0–1.5× | 20–40% | Hold & optimize | Refresh creatives, tighten targeting |
| 0.7–1.0× | 0–20% | Optimize | Pause worst ad sets; fix onboarding |
| < 0.7× | Negative | Pause | Kill channel or cut 50% budget |
Adjust thresholds to user's margin targets from
.
Per-Channel Analysis
For each connected channel, report:
| Field | Detail |
|---|
| Spend (7d) | From platform API |
| Installs | From platform API |
| Conversions | Purchase/subscribe events |
| CPA | Spend / conversions |
| CPI | Spend / installs |
| Revenue | From RC attribution or ASA join |
| ROAS | Revenue / spend |
| Verdict | Scale / Hold / Pause |
Channel-specific notes
| Channel | Caveat |
|---|
| TikTok | iOS data delayed 24–72h; judge at 7 days |
| Meta | AEM modeling on iOS; expect 10–20% discrepancy vs MMP |
| ASA | Most reliable ROAS via Appeeky join |
| Google UAC | Manual tracking; no Appeeky API yet |
Realistic Expectations
Tell the user honestly:
| Topic | Reality |
|---|
| Sustainable margin | 30–50% net after ads + store fees is good for subscriptions |
| Learning spend | First $500 per channel is often unprofitable |
| Geo comparison | Compare CPA to same-geo LTV, not global blend |
| Organic lift | Paid installs often boost organic — true ROAS may be higher |
| Refunds/churn | LTV erodes over time; use 90-day LTV for mature apps |
Scale Readiness Checklist
Before recommending scale on any channel:
Output Template
markdown
# Profitability Report — [App Name] — [Period]
## Unit economics
- LTV: $___ (source: RC / user / estimate)
- Store fee: ___%
- Break-even CPA: $___
- Scale CPA (0.5× net LTV): $___
- Payback period at target CPA: ___ days
## Blended performance
- Total ad spend: $___
- Total attributed revenue: $___
- Blended ROAS: ___×
- Net margin: ___%
- Verdict: Scale / Hold / Pause
## By channel
|---------|-------|----------|------|-----|---------|------|---------|
| ASA | | | | | | | |
| TikTok | | | | | | | |
| Meta | | | | | | | |
| **Total** | | | | | | | |
## Top issues
1. [e.g. "Meta CPA $34 vs $22 target — creative fatigue likely"]
2. [e.g. "TikTok 0 purchase events — MMP mapping broken"]
## Scale recommendations
|---------|--------|--------|------------|
| ASA | Scale +20% | ROAS 1.8×, 14d stable | $/day |
| Meta | Pause | CPA 2.1× LTV | $0 |
| TikTok | Hold | Insufficient data (4 days) | $/day |
## Blockers to profitability
1. [specific blocker with fix]
Cross-Skill Handoffs
| Finding | Route to |
|---|
| ASA keyword detail | |
| Budget reallocation | |
| Channel comparison dashboard | cross-channel-performance
|
| LTV/trial issues | , aso-skills |
| MMP event gaps | |
| Creative fatigue (social) | or |
Related Skills
- — RevenueCat health and LTV baseline
cross-channel-performance
- — budget allocation based on profitability
- — ASA keyword-level profit
- / — refresh fatigued creatives