Getting Started

Overview

conversionflow-aggregate is a sophisticated customer journey analytics platform that combines Bayesian network modelling with genetic algorithm optimisation. This guide will help you get the system running quickly and understand the basic workflow.

Prerequisites

  • Python 3.9 or higher

  • 8GB RAM minimum (16GB recommended for large datasets)

  • 2GB free disk space

  • Git access to the repository

Quick Installation

The fastest way to get started is using the automated setup script:

# Clone the repository
git clone https://github.com/tandpds/conversionflow-aggregate.git
cd conversionflow-aggregate

# Make the pipeline script executable
chmod +x run_pipeline.sh

# Complete automated setup
./run_pipeline.sh --mode=setup

This command will:

  • Create a Python virtual environment

  • Install all required dependencies

  • Validate the installation

  • Run basic system checks

Verification

Test your installation with a quick validation:

# Run integration test (30 seconds)
./run_pipeline.sh --mode=test

If successful, you should see:

✓ ALL TESTS PASSED - System working correctly
✓ Environment validated
✓ Dependencies confirmed

Your First Analysis

Run the Italy market optimisation example:

# Quick optimisation using pre-computed parameters (3-4 seconds)
./run_pipeline.sh --mode=italy

For the complete end-to-end pipeline including Bayesian model fitting:

# Full pipeline with MCMC estimation (7-8 minutes)
python scripts/run_full_italy_pipeline.py

Understanding the Output

The system generates several types of output:

Executive Summary

Professional reporting suitable for business stakeholders, including:

  • Budget allocation recommendations

  • Conservative improvement estimates

  • Implementation guidance

Technical Results

Detailed analytical outputs including:

  • MCMC diagnostics and validation

  • Genetic algorithm convergence metrics

  • Statistical confidence intervals

Visualisations

Charts and diagrams including:

  • Customer journey flow diagrams

  • Budget allocation distributions

  • Performance improvement projections

Next Steps

Common Issues

Installation Problems:

# Clean reinstall if needed
./run_pipeline.sh --mode=setup --force

Memory Errors: Reduce computational requirements in configuration files if running on systems with limited RAM.

Permission Issues: Ensure you have read/write access to the project directory and subdirectories.

For comprehensive troubleshooting, see Troubleshooting Guide.