davidev07/synthetic-ecommerce-dataset

Synthetic E-commerce Dataset - Free Sample

This repository contains a free sample of the Synthetic E-commerce Dataset: 100 users, 50 products, 1,000 transactions, with full relational integrity (foreign keys, temporal consistency, validated business rules) and a fixed random seed (42) for reproducibility.

License below (CC0) applies only to this free sample, not to the full-size commercial packs.

Formats

• CSV
• Apache Parquet

Schema

users: user_id, full_name, email, country, city, currency, signup_date

products: product_id, product_name, category, price

transactions: transaction_id, user_id, product_id, purchase_date, quantity, unit_price, subtotal, discount_percentage, discount_amount, total_amount, currency, country, payment_method, payment_status, order_status, device_type, discount_code, shipping_address

Quick start

import pandas as pd

users = pd.read_parquet("users.parquet")
products = pd.read_parquet("products.parquet")
transactions = pd.read_parquet("transactions.parquet")

result = transactions.merge(users, on="user_id").merge(products, on="product_id")
print(result.head())

Complete Dataset

The commercial package contains:

• 1,000 transactions
• 50,000 transactions
• 500,000 transactions

Each size also includes its own users and products tables. Full-size packs are available for purchase on Gumroad: https://bergese4.gumroad.com/l/synthetic_e-commerce_dataset

All data is completely synthetic. No real customer information is included.