# Customer Segmentation — Test Project Configuration
#
# Data size: 400 rows x 8 columns
# Download: python dev/test-datasets/download.py --dataset marketing
#
# Project name:
marketing-test

# Data path:
/home/mrichardson/Projects/Urika/dev/test-datasets/marketing/data

# Description:
Synthetic customer data simulating a retail company's customer relationship management (CRM)
database. Each row represents a customer with demographic and behavioral attributes:
customer ID, age, gender, annual income, spending score (1-100 composite metric reflecting
purchase value and frequency), education level, marital status, and monthly purchase
frequency. The spending score was designed as a proprietary metric combining transaction
value, frequency, and recency. Income follows a log-normal distribution reflecting real-world
wealth patterns. The research goal is to discover natural customer segments using
unsupervised learning methods (clustering) and to characterize each segment by its
demographic and behavioral profile, enabling targeted marketing strategies.

# Research question:
What natural customer segments exist in the data based on demographic and behavioral
features, and how can each segment be characterized to inform targeted marketing strategies?

# Mode:
exploratory

# Web search:
no

# Venv:
no

# Knowledge suggestions:
Add the data-description.md from dev/test-datasets/marketing/knowledge/. Optionally add
references to RFM analysis or customer segmentation methodology.
