Web User Behaviour Analytics Project

This project simulates and analyses user behaviour on a website. The final output is an interactive dashboard that provides actionable insights for business stakeholders.

Business Value

The goal is to understand user behaviour to drive business decisions regarding:

• Marketing campaign optimisation and timing.
• UX/UI improvements based on device-specific pain points (e.g., high bounce rates).
• User segmentation for personalisation.
• Global expansion strategy by analysing regional behaviour.

---

Technology Stack

• Core Language: Python
• Data Handling: Pandas, NumPy
• Dashboard: Streamlit, Plotly
• Containerization: Docker
• Data Simulation: Faker

---

Project Structure

web-user-analytics/
├── data_simulation/           # Simulated data generator scripts
├── data/                      # Raw and processed data (gitignored)
├── etl_pipeline/              # ETL scripts for processing data
├── analysis/                  # EDA notebooks / scripts
├── dashboard/                 # Streamlit dashboard app
├── Dockerfile                 # Docker container definition
├── .dockerignore              # Files to ignore in Docker build
├── requirements.txt           # Python dependencies
└── README.md                  # Project description

---

How to Run

There are two ways to run this project: using Docker (recommended) or setting up a local Python environment.

Option 1: Run with Docker (Recommended)

This is the easiest way to run the entire application.

Prerequisites:

• Docker must be installed and running on your system.

Steps:

1. Build the Docker image: Open your terminal in the project root and run:

   docker build -t web-analytics .

2. Run the Docker container:

   docker run -p 8501:8501 web-analytics

3. View the Dashboard: Open your web browser and navigate to http://localhost:8501. The dashboard should be live.

Option 2: Run Locally

Prerequisites:

• Python 3.8+

Steps:

1. Clone the repository.

2. Set up a virtual environment (recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows, use `venv\Scripts\activate`

3. Install dependencies:

   pip install -r requirements.txt

4. Run the pipeline and dashboard: To see the dashboard, you must first generate and process the data.

   python data_simulation/generate_data.py
   python etl_pipeline/process_data.py
   streamlit run dashboard/app.py

---

Example Insights

The analysis of this simulated data typically reveals several key insights:

• Desktop users convert at a higher rate than mobile or tablet users, suggesting the desktop experience is more conducive to purchasing.
• Mobile users have the highest bounce rate, indicating potential UX friction or performance issues on the mobile site.
• User engagement and session volume peak in the evening, which could be the optimal time for marketing pushes and announcements.
• Session length and pages visited are strong positive indicators of a user's likelihood to convert.