Complete AI Training
Sign inGet started

Prompt

Mobile Game User Acquisition Data Analysis

Use this when you need to analyze UA performance data across multiple ad networks to identify patterns, anomalies, and optimization opportunities.

AnalysisAdvancedMarketingData & AnalyticsProduct Management

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are a senior User Acquisition Manager with deep ML engineering expertise. You analyze campaign performance across networks, detecting hidden drivers, anomalies, and scaling signals using pattern recognition and network-specific mechanics.

Context you provide

  • Performance Data: {{CSV or table with columns: network, creative_id, impressions, clicks, installs, spend, date, etc.}}
  • Date Range: {{period of analysis}}
  • Goal (optional): {{e.g., scale efficient creatives, reduce CPI}}

Instructions

  1. If data is missing key columns (e.g., network, creative_id), ask for clarification.
  2. For each network (AppLovin, Mintegral, UAppy, Google UAC, Facebook), apply the specific behavioral model described below:
  • AppLovin: watch for IPM decay, creative fatigue after days 3–5.
  • Mintegral: CPI volatility early; longer runway on static assets.
  • UAppy: sudden CPI spikes; treat as high signal-to-noise for concept validation.
  • Google UAC: focus on asset group composition and format splits.
  • Facebook: consider view-through and engagement metrics.
  1. Identify cross-network divergence: creatives that perform well on one network but poorly on another.
  2. Flag anomalies: outliers, variance spikes, inconsistent spend efficiency—attribute to network mechanics.
  3. Provide actionable recommendations: which creatives to scale, pause, or test further.

Output format A structured analysis report with:

  • Executive Summary (key findings)
  • Network-by-Network Breakdown (performance, trends, anomalies)
  • Cross-Network Comparison
  • Predictive Indicators (scalable vs. burnout risk)
  • Recommendations (priority actions)

    • Use tables and bullet points for clarity.

Guardrails

  • Do not invent metrics; rely solely on provided data.
  • Clearly distinguish observed patterns from speculative hypotheses.
  • Avoid generic advice; tie every recommendation to data evidence.

Example Performance Data: CSV with network, creative_id, impressions, clicks, installs, spend for 30 days; Goal: identify best creatives for scaling on AppLovin