This AI Marketing Automation System automates marketing campaign management by analyzing performance data, applying predefined optimization rules, and leveraging AI-powered insights. It supports budget adjustments, campaign pausing, and other decisions while providing actionable insights and visualizations to enhance marketing effectiveness.

• Data Ingestion: Upload campaign performance data from CSV files.
• Performance Analysis: Evaluate campaigns using key metrics like CTR (Click-Through Rate), CPA (Cost Per Acquisition), and ROAS (Return on Ad Spend).
• Optimization Rules:
  • Pause campaigns with poor CTR or high CPA.
  • Increase budgets for campaigns with excellent ROAS.
  • Decrease budgets for underperforming campaigns.
• AI Insights: Generate suggestions for ad creative improvements and targeting strategies.
• Visualization:
  • Parallel coordinate plots for campaign comparison.
  • ROI bubble charts to visualize performance efficiency.
  • Revenue waterfalls to track profitability.
• Reporting: Generate PDF reports summarizing actions and campaign performance.
• Database Integration: Store and retrieve historical campaign data and recommendations.

project-root/
├── app.py                  # Main Streamlit application
├── requirements.txt        # List of dependencies
├── README.md               # Documentation
├── data/                   # Datasets and database files
│   ├── synthetic_campaign_data.csv  # Sample input data
│   └── db.py               # Database setup and interaction
├── functions/              # Modular functions for processing
│   ├── load_data.py        # Data loading and preprocessing
│   ├── batch_processing.py # Batch processing campaigns
│   ├── generate_insights.py # AI insights generation
│   ├── action_executor.py  # Action execution logic
│   ├── visualization.py    # Advanced visualization functions
│   └── reporting.py        # PDF report generation
└── outputs/                # Generated outputs (e.g., PDF reports)
    └── campaign_report.pdf

git clone <repository_url>
cd project-root

pip install -r requirements.txt

streamlit run app.py

• Upload a CSV file containing campaign performance data with the following columns:
  • Campaign ID: Unique identifier.
  • Impressions: Number of times an ad was shown.
  • Clicks: Number of clicks received.
  • Conversions: Number of successful actions (e.g., purchases, signups).
  • Spend: Total spend in currency.
  • Revenue: Total revenue generated.
  • Status: Current status (Active, Paused).

• Set the following optimization thresholds:
  • Target CPA: Maximum acceptable cost per acquisition.
  • Minimum CTR: Minimum acceptable click-through rate.

• Click Run Analysis to process campaigns.
• View recommendations such as "Pause due to low CTR" or "Increase budget due to high ROAS."
• Visualize campaign performance trends in multiple formats.

• AI-powered insights will suggest ad creative improvements and targeting strategies.

• Click Generate PDF Report to create a detailed summary of actions and metrics.

The system includes a sample dataset: data/synthetic_campaign_data.csv.

Sample Rows:

• CTR: Clicks ÷ Impressions
• CPA: Spend ÷ Conversions
• ROAS: Revenue ÷ Spend

• Pause Campaigns:
  • CTR < Minimum CTR
  • CPA > 3 × Target CPA
• Increase Budget:
  • ROAS > 4
• Decrease Budget:
  • ROAS < 1.5

• Comparison Chart: Side-by-side metric comparison across campaigns.
• ROI Bubble Chart: Visualize return on investment for campaigns.
• Parallel Coordinates: Analyze relationships between metrics.
• Revenue Waterfall: Track campaign profitability step-by-step.

• Python: Core programming language.
• Streamlit: Interactive web interface.
• Pandas: Data manipulation.
• Plotly: Advanced visualizations.
• FPDF: PDF report generation.
• SQLite: Database integration for storing history.
• LLMs: Generate insights using GPT-based APIs.

• Use platforms like Google Ads API or Meta Ads API for real-time data ingestion and execution of actions.

• Train custom models to predict campaign performance and provide more detailed optimization suggestions.

• Deploy on cloud platforms (AWS, GCP) to handle large-scale datasets and concurrent users.

• Extend visualizations with real-time interactive dashboards using tools like Dash or Tableau.

{
  "campaign_id": "CAMP_041",
  "current_status": "Paused",
  "current_status_reason": "Paused due to low CTR: 0.58% (Below 1.00%)",
  "action": "PAUSE",
  "reason": "Low CTR: 0.58% (Below 1.00%)",
  "metrics": {
    "CTR": 0.0058,
    "CPA": 45.32,
    "ROAS": 1.12
  }
}

The PDF report includes:

• Campaign recommendations.
• Metrics like CTR, CPA, ROAS.
• Actions (e.g., Pause, Increase Budget).

This project is licensed under the MIT License.